|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001441498
007 cr mnu|||uuuuu
008 031203s2003 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000171
Todd, Melissa Farino.
Perceptions of family functioning between children with behavior difficulties and their primary caregiver
h [electronic resource] /
by Melissa Farino Todd.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ed.S.)--University of South Florida, 2003.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 124 pages.
ABSTRACT: This research study compared perceptions of family functioning among preadolescent children with behavior difficulties and their primary caregivers. Participants consisted of 29 caregiver-child dyads as well as each child's classroom teacher. Eligibility for the study was based on the child's placement within a self-contained Emotionally Handicapped (EH) or Severely Emotionally Disturbed (SED) classroom in one of three elementary schools within two west coast Florida counties. Data collection included teacher rating scales pertaining to the severity of each child's behavior and the presence of Callus Unemotional (CU) traits in addition to caregiver and child interviews tapping perceptions of family functioning. Results indicated that caregivers consistently viewed their families as more adaptive and cohesive than did children with a disruptive behavior disorder. These findings are consistent with previous research showing a similar pattern among older adolescents with a disruptive behavior disorder. No relationship was not found between the child's perception of family functioning and CU traits, although it was noted that there was considerable restriction of range on CU traits. Overall, the results of this study contributes to the existing literature by demonstrating that preadolescents, like their older counterparts, also view their families as less adaptive and cohesive than do their caregivers. Limitations and directions for future research are discussed.
Adviser: Mendez, Linda Raffaele
oppositional defiant disorder.
x School Psychology
t USF Electronic Theses and Dissertations.
Perceptions of Family Functioning Between Children with Behavior Difficulties and their Primary Caregiver by Melissa Farino Todd A thesis submitted in partial fulfillment of the requirements for the degree of Educational Specialist Department of Psychological and Social Foundations College of Education University of South Florida Major Professor: Linda Raffaele Mendez, Ph.D. George Batsche, Ed.D. John Ferron, Ph.D. Date of Approval: October 31, 2003 Keywords: Externalizing Disorders, Conduct Disorder, Oppositional Defiant Disorder, Children, Callous-Unemotional Traits Copyright 2003, Melissa Farino Todd
i Table of Contents List of Tables .. v List of Figures vi Abstract vii Chapter I: Introduction 1 Statement of the Problem ... 4 Purpose of Study... 4 Definitions .. 4 Research Questions ... 5 Exploratory Questions.. 5 Importance of the Current Study .. 6 Chapter II: Review of the Literature 7 Disruptive Behavior Disorders: Definitions and Etiology ... 7 Oppositional Defiant Disorder...... 7 Conduct Disorder .. 8
ii Demographic Variables Associated with Disruptive Behavior Disorders..... 10 Socioeconomic Status .. 11 Maternal Adjustment ... 12 Family Interactions Associated with Disruptive Behavior Disorders. 13 Family Discord .. 14 Coercive Family Process .... 16 Parental Attitudes and Beliefs ... 19 Parental Management of Behavior ... 21 Family Systems Theory 27 The Circumplex Model of Marital and Family Systems 29 Perceptions of Family Functioning ... 31 Summary .. 37 Chapter III: Method 40 Participants .. 40 Measures 41 Disruptive Behavior Disorders Rating Scale Teacher Form 41 Family Adaptability and Cohesion Evaluation Scale ... 43 Family Adaptability and Cohesion Evaluation Scale Children. 45 Psychopathy Screening Device .. 45 Procedure .... 49
iii Pilot 50 Study .. 51 Research Design and Statistical Analysis .. 53 Chapter IV: Results 55 Demographic Data ... 55 Internal Consistency Reliability Estimates ... 57 Marginal Means and Repeated Measures ANOVA .... 59 Correlation Matrix .... 66 Chapter V: Discussion 71 Differences in Caregiver and Child Perception of their Familys Family Functioning .... 71 Childs Perception of Family Functioning and Callous-Unemotional Traits ... 75 Practical Implications and Contributions to the Literature ..... 77 Limitations and Directions for Future Research ..... 78 References .. 83 Appendices Appendix A: Family Demographic Information Sheet.... 90
iv Appendix B: Disruptive Behavior Disorders Checklist 91 Appendix C: Family Adaptability and Cohesion Evaluation Scales Parent Version 95 Appendix D: Family Adaptability and Cohesion Evaluation Scales Childrens Version .. 97 Appendix E: Psychopathy Screening Device.. 108 Appendix F: Parent Letter... 110 Appendix G: Consent Form... 111 Appendix H: Child Assent...... 112 Appendix I: Group Instructions...... 113 Appendix J: Debriefing Protocol.... 114 Appendix K: Thank You Letter ..... 115
v List of Tables Table 1. Participants Demographic Characteristics .. 56 Table 2. Cronbachs Alpha for the Rating Scales 58 Table 3. Descriptive Statistics for Family Functioning Measure 59 Table 4. Descriptive Statistics for the FACES Categories 60 Table 5. Descriptive Statistics for the Cohesion Subscale of the FACES 62 Table 6. Descriptive Statistics for the Adaptability Subscale of the FACES. 64 Table 7. Correlations Across Raters and Responses on the FACES 67 Table 8. Correlations Between CU Traits and Childs Perception of Family Functioning... 68
vi List of Figures Figure 1. FACES Mean Scores for Child and Caregiver 65 Figure 2. Scatterplot of Parent and Child Responses on the Cohesion Subscale of the FACES.. 66 Figure 3. Scatterplot of Parent and Child Responses on the Adaptability Subscale of the FACES 66
vii Perceptions of Family Functioning Between Children with Behavior Difficulties and their Primary Caregiver Melissa Farino Todd ABSTRACT This research study compared perceptions of family functioning among preadolescent children with behavior difficulties and their primary caregivers. Participants consisted of 29 caregiver-child dyads as well as each childs classroom teacher. Eligibility for the study was based on the childs placement within a self-contained Emotionally Handicapped (EH) or Severely Emotionally Disturbed (SED) classroom in one of three elementary schools within two west coast Florida counties. Data collection included teacher-rating scales pertaining to the severity of each childs behavior and the presence of Callus Unemotional (CU) traits in addition to caregiver and child interviews tapping perceptions of family functioning. Results indicate that caregivers consistently view their families as more adaptive and cohesive than do children with a disruptive behavior disorder. These findings are consistent with previous research showing a similar pattern among older adolescents with a disruptive behavior disorder. No relationship was not found between the childs perception of family functioning and CU traits,
viii although it was noted that there was considerable restriction of range on CU traits. Overall, the results of this study contributes to the existing literature by demonstrating that preadolescents, like their older counterparts, also view their families as less adaptive and cohesive than do their caregivers. Limitations and directions for future research are discussed.
1 Chapter I Introduction Disruptive behavior disorders such as Conduct Disorder (CD) and Oppositional Defiant Disorder (ODD) are serious and pervasive problems for the diagnosed child as well as for his or her family. These externalizing disorders affect individuals in multiple domains of their lives including home, school, and community. Children with ODD or CD tend to lack the problem-solving skills necessary to respond appropriately to situations, and thus they often respond in a hostile or defensive manner. Employing such maladaptive approaches to problem-solving over time strengthens an individuals behavioral repertoire such that these responses eventually become prepotent. Research has shown that no treatment or ineffective treatment for ODD and/or CD fosters a negative trajectory leading to antisocial behavior in adulthood. Consequently, it is imperative to develop a thorough understanding of these disorders and the factors that contribute to them in an effort to prevent later maladaptive outcomes. Although the literature identifies several factors related to the etiology of disruptive behavior disorders, it is clear that deviant behavior is multidetermined, making distinct conclusions about causation impossible. However, a recurring theme in the research is that the family environment and interaction among family members serve as a possible precursor to ODD or CD. For example,
2 family discord (Abidin et al., 1992; Rutter, 1985; Dadds, 1987; Frick, 1993; Rutter, 1994; & Gardner, 1992), parental attitudes and beliefs (Baden & Howe, 1992, Haddad et. al., 1991), and behavior management strategies (Gardner, 1992; Rutter, 1985; Gelfand et. al., 1982) have all been suggested as contributors to a maladaptive family system. More specifically, Patterson (1997) proposed that CD emerges as a result of intense and constant negative interactions between parent and child. This research suggests that a family systems approach to the understanding and treatment of disruptive behavior disorders is essential. From a theoretical standpoint, a family systems approach considers each individual family member as part of a whole system, with each individual affecting the behavior of the other. The Circumplex Model of Marital and Family Systems is based on family systems theory and provides a descriptive means by which to examine various types of family relationships. Three key components of family functioning that are assessed within this model are the degree of balance a family possesses in regard to their emotional connectedness (cohesion), flexibility in family roles/rules (adaptability), and communication style. The core of the Circumplex Model is the focus on perceptions of family functioning according to various family members. Research utilizing this model has found that parents often differ in their perceptions of family functioning when compared to their child with behavior disorders (Nollar & Callan, 1986; Nollar, Seth-Smith, Bouma, & Schweitzer, 1992; Slee, 1996), with children perceiving their families as less adaptive and cohesive than their parents. In addition, Pillay (1998)
3 revealed differences in perceptions of family functioning between adolescents with CD and their non-disordered peers. Specifically, his results indicated that adolescents with CD viewed their families as less adaptive and cohesive, and were generally less satisfied with their familys functioning. These studies indicate that adolescents with ODD or CD tend to view their families as being emotionally unsupportive and rigid. Whether such findings reflect a hostile, negative bias toward the family among adolescents with ODD or CD or an accurate picture of the familys functioning is unclear at this time. It also is unclear if young children with ODD or CD view their families as negatively as do adolescents with these disorders, The exclusion of preadolescents in exploring perceptions is a limitation of the aforementioned studies, inhibiting our understanding of how the perceptions of younger children with ODD or CD differ from the perceptions of other family members. Given that family functioning is purported to be a factor contributing to the development of disruptive behavior disorders in children, an important question that arises in considering the family is why some children in the family develop ODD or CD and others do not. One premise is that there are child factors that are independent of, or interact with, family functioning that play a role in the deviant behavior. According to Wooton, Frick, Shelton, and Silverthorn (1997), these factors consist of a temperament style laden with emotional constrictedness, lack of guilt, and lack of fearful inhibitions, resulting in a decreased responsiveness to punishment and an increased resistance to parental and societal norms. In examining the role of these traits (termed callus
4 unemotional traits or CU traits), Frick (2000) found that children high on CU traits seem to develop CD independent from family functioning, concluding that family functioning is more strongly associated with CD in children without CU traits. Statement of the Problem Although the literature addresses the issue of family functioning in relation to ODD and CD, the focus is on family relationships within a specific developmental timeframe (i.e., mother and adolescent) as opposed to the perceptions of preadolescents. Further, there are no studies that examine CU traits as they relate to family members perceptions of family functioning. The current study expanded upon the existing literature by including the perceptions of the preadolescent child with ODD or CD in comparison to their primary caregiver. Further, CU traits were examined in relation to perceptions of family functioning to determine the relationship between these variables. Purpose of Study The purpose of this study was to (1) determine the perceptions of family adaptability and cohesion among primary caregivers and preadolescent children diagnosed with ODD or CD, (2) examine the differences between the perceptions of the two groups, (3) identify the level of callous-unemotional traits within children with ODD or CD, and (4) examine the relationship between the childs perceptions of family functioning and CU traits. Definitions 1. Child with a Disruptive Behavior Disorder A child between the ages of seven and twelve who scored 2 standard deviations above the mean as
5 compared to the normative sample on the Disruptive Behavior Disorders Checklist Teacher Rating Scale (DBD-TRS). 2. Primary Caregiver An adult living in the same home as the child who is primarily responsible for raising the child (e.g., biological parent, foster parent, adoptive parent, grandparent, aunt, uncle, etc.). Research Questions 1. How do children with ODD or CD rate their familys adaptability and cohesion? 2. How do primary caregivers of children with ODD or CD rate their familys adaptability and cohesion? 3. Are there differences in perceptions of family adaptability and cohesion between primary caregivers and the identified children with a disruptive behavior disorder? Exploratory Questions 1. What percent of children with ODD or CD in this sample rate high on CU traits? 2. What is the relationship between CU traits and child perceptions of (a) family cohesion, and (b) family adaptability?
6 Importance of the Study There is a strong potential for childhood behavior problems to continue along a trajectory that may lead to adult antisocial behavior. Thus, there is a tremendous need to further understand disruptive behavior disorders, how they develop, and the interactions that potentially maintain them. The current study focused on the differences in perceptions between children with behavioral problems and their caregivers. Dyadic perceptions were obtained to examine how the primary caregiver experienced the family as compared to the child with behavioral issues. Notably, this study used a sample of elementary school aged children and their primary caregivers, thus expanding beyond previous studies that have examined adolescent-parent dyads. In studies such as this one, it is important to note that the accuracy of the participants perceptions cannot be determined since the method does not include objective evaluation of the family (e.g., ratings of family functioning by an independent observer). Rather, the reality of each family member was obtained through a measure of perceptions, which helps to clarify how different family members view the family. In this case, however, perceptions may be as important as objective reality in that they reveal differences in how family members perceive each other and the family as a whole. Furthermore, it appears logical that individuals respond to their environment based on their perception of reality, thus providing additional merit to the importance of obtaining this unique point of view.
7 Chapter II Literature Review Disobedient and non-compliant behavior is found among children and adolescents of all family backgrounds and dynamics. Although some degree of noncompliance is found in many children, for some families, their childs defiance becomes so frequent and extreme that it dramatically affects the home environment, school life, and all relationships involved. In these cases, children may be identified as having Oppositional Defiant Disorder (ODD) or, in more extreme cases, Conduct Disorder (CD). Disruptive Behavior Disorders: Definitions and Etiology Oppositional Defiant Disorder The main diagnostic criterion for ODD according to DSM IV TR (2000) is a pattern of negativistic hostile and defiant behavior lasting at least 6 months. At least four of the following behaviors must be present during that period: (1) loses temper, (2) argues with adults, (3) actively defies or refuses to comply with adults request or rules, (4) deliberately annoys others, (5) blames others for own mistakes or misbehavior, (6) touchy or easily annoyed, (7) angry and resentful, and (8) spiteful or vindictive. In addition, the criteria require that the disturbance in behavior cause clinically significant impairment in social, academic, or occupational functioning.
8 Conduct Disorder. The essential feature in the diagnosis of CD is a repetitive and persistent pattern of behavior that violates the basic rights of others and major age appropriate societal norms (DSM-IV-TR, 2000, American Psychiatric Association). The diagnosis of Conduct Disorder requires three or more of the following behaviors to be present during the past twelve months with at least one present in the past six months: (1) aggression toward people and animals, (2) destruction of property, (3) deceitfulness or theft, and (4) serious violations of rules. Examples of such behavior include: bullying, truancy, running away, fire setting, car theft, substance abuse, and prostitution. The disturbance in behavior can be specified as mild, moderate, or severe depending on the severity of the problems and must cause clinically significant impairment in social, academic, or occupational functioning. Conduct Disorder has been identified as one of the most common forms of psychopathology in children and adolescents (Steiner, 1997), with the prevalence rate estimated between 1.5% and 10% (Mash & Barkley, 1996; Short & Shapiro, 1993). Although boys are more frequently diagnosed with CD in childhood, the gap between boys and girls narrows as children enter adolescence. Research indicates that boys with CD exhibit more overt behaviors (i.e., fighting) while girls with CD engage in more covert behaviors (i.e., lying, cheating, shop lifting, and truancy; Mash & Barkley, 1996; Steiner, 1997; American Psychiatric Association, 2000). The onset of CD typically begins in early childhood and extends into adulthood with three key factors differentiating it from other childhood problems and behaviors, namely antisocial behavior, chronicity, and impairment in
9 functioning (Short & Shapiro, 1993). In other words, a child with CD displays behavior that is antisocial, continuous, and causes impairment in functioning in at least one important domain (e.g., home, school, community). It is important to note the proposed trajectory from ODD to CD. Specifically, ODD tends to emerge approximately three years prior to the typical manifestation of CD and is suggested to be a mild precursor for CD (Mash & Barkley, 1996). Despite this proposed progression, Mash and Barkley (1996) note that while 90% of individuals with CD have previously met the criteria for ODD, the majority of the children diagnosed with ODD do not develop the more severe behaviors characteristic of CD. The literature pertaining to these disorders shows that children diagnosed with ODD or CD lack effective problem-solving skills (Dodge, 1993, Hemphill, 1996, Barkley, 1998, & Pillay, 1998), resulting in poor relationships with others. According to Dodge (1993), there are five specific areas in which aggressive children have deficits including: (a) difficulties attending to and perceiving information, (b) biased and inadequate interpretations of social cues, (c) tendency to recall more hostile social cues from memory, (d) generation of fewer steps to solving a problem and fewer possible obstacles, and (e) the tendency to chose an aggressive rather than prosocial response as a solution. As children engage repeatedly in these maladaptive approaches to social situations, they strengthen the cognitive structures that support these behaviors, which enables their aggressive behavior to become automatic (Dodge, 1993).
10 A number of factors are associated with the occurrence of disruptive behavior disorders, including biological, sociocognitive, peer/community, and familial factors (Steiner & Wilson, 1999; Ghuman, 1998; Mash & Barkley, 1996, Short & Shapiro, 1993). The literature provides evidence that deviant behavior is multidetermined (Mash & Barkley, 1996), meaning that there is no clear separation of the potential factors that contribute to the development of ODD and/or CD. Additionally, it is well established that given the maladaptive outcomes associated with disruptive behavior disorders, there is a dire need for effective interventions. No treatment or ineffective treatment typically results in a predicable negative trajectory that can be explained by the accumulation of risk throughout the lifespan (Steiner, 1997). An influential domain that may serve as both a contributor to the manifestation of a DBD as well as a target for intervention is the family. Several family variables, including demographic variables and family interaction variables, have been found to be associated with the presence of ODD and/or CD in children and will be explored in this review of the literature. Demographic Variables Associated with Disruptive Behavior Disorders The types of demographic variables most pertinent to the research on families with a child diagnosed with ODD or CD are socioeconomic status and maternal adjustment. Although many of these variables are correlational rather than causal, they are important because they contribute to the understanding and treatment of families with a child diagnosed with a DBD.
11 Socioeconomic status One of the most commonly identified demographic family variables that is related to behavior problems in children is low socioeconomic status (SES). Although low SES does not cause severe behavior problems, numerous studies (Frick et al., 1989; Haddad et al., 1991; Rutter, 1985; Behar & Stewart, 1984) have found that this characteristic is associated with the occurrence of ODD and CD. It is important to note, however, that it is not low SES alone, but low SES in combination with other variables such as maternal antisocial personality, low family cohesion, and high family conflict (Frick et al., 1989) that is associated with the development of DBDs. This finding suggests that low SES may be a mediating variable in that socioeconomic disadvantage places the child at higher risk for the development of ODD and/or CD when low SES is combined with other variables (e.g., parental discord, aversive parent-child interactions). Due to the strong interconnected relationship between these variables, a causal relationship between SES and childhood DBDs cannot be assumed. In examining socioeconomic status as a risk factor contributing to the development of ODD and/or CD, McGee and Williams (1999) suggested several potential trajectories. First, they suggested that the persistent poverty experienced by low SES families places an extraordinary amount of stress on parents, resulting in an interference in parenting skills. Relatedly, Haddad (1991) noted that the parental values of low SES families might contribute to the high incidence of aversive behaviors among their children. In comparing high SES parents to low SES parents, Haddad noted that the former emphasized an
12 internalized system of self-direction whereas the latter emphasized conformity to externally imposed rules. These differences in disciplinary styles are significant in the acquisition of values and behavior. Second, the lack of a significant income limits a familys access to health care, which hinders the probability of receiving effective treatment. Lastly, children from low SES homes are more likely to be exposed to unsafe or unhealthy environments. Such environments may include a range of negative situations, from witnessing physical violence at home or in the community to lack of supervision and parental support. Maternal adjustment Another demographic variable that has been linked to disruptive behavior disorders is maternal adjustment. Abidin (1992) suggested that social competence, adaptational competence, and self-esteem are maternal attributes that are related to maternal adjustment and coping with children. In addition to these internal characteristics, there are environmental factors, such as lack of support that may have a negative effect on maternal adjustment. Abidin (1992) purported that a lack of social and spousal support results in inappropriate parenting behaviors by mothers, as well as the development of impaired relationships with their children. These contributing maternal characteristics can be challenging and can elicit negative moods in normal mothers who have a non-problem child; therefore, it is plausible that these effects are even greater in families who have a child identified as having a disruptive behavior disorder. Much of the research linking maternal characteristics with child behavior problems reviews the repercussions of maternal depression. Dadds (1984)
13 found that children of depressed mothers have significantly more emotional, somatic, and behavioral problems than children of nondepressed mothers (pg. 348). The hypotheses concerning the nature of the relationship between maternal depression and child behavior problems note that depressed mothers (1) appear to have a perceptual or cognitive bias such that they rate their childs behavior as deviant or aversive when, to independent observers, the child is behaving in a neutral manner, and (2) may engage in relatively low frequencies of positive interaction with their children and be more inconsistent in their use of discipline. Although Dadds (1994) proposed that depression influences mothers to rate their children more negatively on behavior rating scales, research has been conducted that contradicts this theory. Specifically, Gardner (1992) has shown that the utilization of home observations and father reports indicates that both depressed mothers and non-depressed fathers produced scores that were similar to those of non-depressed mothers when rating a particular child on degree of difficulty. These results suggest that depressed mothers do not imagine that their children are difficult and that the interplay between these variables is likely quite complex. Family Interactions Associated with Disruptive Behavior Disorders In addition to examining the influence of particular demographic variables on the development of ODD and/or CD, research has addressed the impact of family interactions on the development of these disorders. Examples of such
14 interactions include family discord, child-parent attachment, parental attitudes and beliefs, and parent management of child behaviors. Family discord One common interactional pattern found in families with a child diagnosed with a DBD is family discord. Family discord is defined as disharmony among family members, which may or may not directly involve the child. The presence of family discord can be detrimental to normal child development and adjustment. Considerable evidence indicates that parental conflict is a strong predictor of behavioral problems among children (Abidin et al., 1992; Rutter, 1985; Dadds, 1987; Frick, 1993; Rutter, 1994; & Gardner, 1992). Dadds (1987) noted that children, whether from divorced or intact families, are at a greater risk to develop behavior problems when exposed to open marital discord. In addition, it has been speculated (Abidin, 1992) that parental conflict results in less sensitive parenting, which may be a major factor underlying childrens adjustment problems. Rutter (1994) discussed several alternatives in an effort to determine which aspect of family discord can be considered the source of risk for children in regard to the development of ODD and/or CD. He noted that the risk could derive from children: (a) witnessing strongly negative interchanges between parents, (b) being drawn into the marital discord or parent/child discord, (c) experiencing the lack of a supportive, affectionate relationship between the parents or between themselves and their parent, (d) receiving differential treatment or favoritism from their parents, (e) becoming a scapegoat for the family problems, or (f) being the victim of maladaptive parenting practices.
15 These aforementioned possibilities shed light on the numerous variables that need to be considered when examining the impact of marital discord on the development of a DBD. It is apparent that many children who develop ODD or CD come from families who engage in relatively high rates of aggressive, coercive behaviors on a daily basis. This is supported by Dadds (1997), who noted that the families of oppositional children contribute to the childs assimilation of aggressive behaviors by providing them with a coercive environment. It is important to recognize, however, that research has not established the direction of the relationship between childhood behavior disorders and family discord (Hemphill, 1996). In other words, the childs coercive behavior may be so severe that it places a strain on family dynamics (i.e., marital subsystem, sibling subsystem) causing greater family conflict, or it may be the case that there is pre-existing conflict within the family that disrupts the childs functioning. Rutter (1994) has suggested that family discord impairs the security of childrens attachment relationships, which predisposes a child to develop Conduct Disorder. Secure attachment relationships, according to Robinson (1985), produce children who are more socially competent, have more friends, are more empathic, and are more capable of reciprocity. Therefore, the residual effects of parental conflict hinder the appropriate development of a secure attachment. Without this opportunity to establish a secure attachment, children are placed at risk for developing an interaction pattern that is aggressive and
16 self-centered, which is characteristic of children and adolescents with psychopathology (Robinson, 1985). Coercive family process Patterson and his colleagues (1997) have identified a pattern of interaction between parents and children that they have termed coercive family process. This process, which characterizes many families with children with behavior problems, serves as the training ground for the development of antisocial behavior. Coercive antisocial behavior has been defined as contingent aversive behavior, where the behaviors that contingently follow certain classes of events are examined to determine the function of antisocial behavior within social interaction or behavioral contexts (Dishion, French & Patterson, 1995). Patterson, Dishion, and Banks (1984) noted that patterned irritable exchanges between a child with behavior problems and other family members serve as the basic training for coercion. The first phase of the training is the continuous failure of the parents to use effective discipline in controlling the coercive behaviors between family members (Patterson, Dishion, & Banks, 1984). The pattern of coercive exchanges therefore increases in frequency and severity and leads to further disruption in parental efforts to discipline. Thus, the child learns antisocial behavior within parent-child exchanges. The causal arrows between parent and child in this model go both ways, meaning that the more coercive the child becomes, the more difficult he/she is to manage. The primary interactional pattern that has been suggested to contribute to the childs coercive behavior is negative reinforcement, which in this theory is the
17 process by which the child learns to avoid parent demands. More specifically, the continuous presentation of negative reinforcement results in the child learning to use coercive behaviors to gain control of the family environment (Dishion, French & Patterson, 1995). An example of this coercive trap is as follows: (1) the child exhibits aggressive or aversive behavior, (2) the parent attempts to punish the child is some way, (3) the child persists in the behavior or rebels, (4) the parent withdraws punishment (Combrinck-Graham, 1990). This sequence confirms to the child that persistence results in gaining control and eventually succeeding in getting what he or she wants. Lyons-Ruth (1996) and Rutter (1994) noted that Patterson and Banks model of early starters illustrates the pattern of development over time typically seen in families with inconsistent or insecure attachment relationships. The cycle is a progressive, three step model that begins with an early coercive interaction between parent and child, characterized by scolding and explosive, irritable, and inconsistent discipline (step 1). This interaction leads to escalating child aggressive behavior, which, in step 2, produces peer rejection, failure in school, and depressed mood. These developments at Step 2 are followed, in turn, by increased involvement in delinquent acts, deviant peer groups and substance abuse, as well as failures at work. According to Lyons-Ruth (1996), there are particular characteristics that put families at risk to engage in the coercive process. For example, lack of social competence and antisocial traits as displayed by the parent can facilitate the development of inconsistent attachment relationships as well as provide the child
18 with a model of aggressive behavior. In addition, a child with a difficult disposition contributes to parental responses, which, in turn, may be negative. Families with high levels of negative interaction are likely to develop escalating cycles of reciprocated aggression (Baden & Howe, 1992). As these negative cycles of family interaction continue, children are provided with the opportunity to increase the intensity and aversiveness of their behavior. Over time, these behaviors become overlearned and automatic, operating without conscious, cognitive control (Dishion, French & Patterson, 1995). In the absence of effective disciplinary practices, the childs aversive behavior in the home may generalize into other settings as the child demonstrates similar patterns in his interactions with others. In a sense, the more difficult the child is, the more control he/she has over his/her immediate surroundings, providing that appropriate interventions are not put into place. The underlying theme in coercive family process is that the parents use inconsistent, harsh, or erratic efforts to set limits for their child (Dishion, French & Patternson, 1995). Interestingly, parents in these circumstances tend to believe that they are using good parenting practices and that the child is just failing to respond. Patterson, Reid and Dishion (1992) found that in contrast to the belief that parents of children with ODD or CD are constantly engaging in negative interactions with their child, only about 10% of the interactions they observed between child and parent were aversive, with the remaining 90% being positive or neutral. However, these authors noted that these 10% of interactions have a much greater impact on the childs development than the other 90% of the
19 interactions. More specifically, Patterson, Dishion & Bank (1984) suggested that aggressive child behaviors are maintained by mixed schedules of positive and negative reinforcement plus punishment. So, even though aversive interactions occur only one tenth of the time, these aversive interactions (many of which consist of negative reinforcement) serve to strengthen the childs coercive reactions. Parental attitudes and beliefs In addition to the type of attachment formed between parents and their children and the perpetuation of coercive family processes, the attitudes and beliefs maintained by parents contribute to the development of a childs deviant behavior. For example, Baden and Howe (1992) expanded upon Pattersons coercion model and proposed an alternative set of parental perceptions that are implicated in coercion cycles. The first involves parent attributions about the causes of their childrens actions. This hypothesis assumes that parents who view their childs negative behavior as internally caused and intentional will be more likely to initiate negative interactions with their child. If this hypothesis is true, then one would expect coercion cycles to be more likely to develop in families in which parents believe their children are solely responsible for their own misbehavior rather than that the behavior is influenced by parental actions. The second set of beliefs involves parental attributions regarding the stability, globality, and controllability of child misbehaviors and their own expectancies regarding their ability to manage their children (Baden & Howe, 1992). More specifically, this belief expands upon parents perceptions about the
20 cause of their childs behavior to how they view their ability to manage their child. In this case, coercion cycles would be most likely be present in families in which parents believe the childs behavior is not due to parental factors but to global and stable attributes of the child. More specifically, the belief that the childs behavior is not controllable by the parent reduces the parental expectancies regarding their ability to manage their child. Consequently, the reduced expectancies for personal effectiveness turn into a state of learned helplessness in which the parent withdraws or backs down in the face of conflict with their child, thus strengthening the coercive process. In testing the aforementioned hypotheses, Baden & Howe (1992) found that parents of children with CD do attribute their childs behavior to child intent that is due to stable, global factors that are outside of parental control. Further, the findings indicated that these parents expect that attempts to influence their childs behavior will be ineffective. However, they did note that these findings do not reveal whether such parental cognitions of blame and helplessness are precursors of coercion cycles or whether they emerge as a result of already established cycles. Parental perceptions of child development and attributions about children also influence the development of children. Haddad et al. (1991) noted that parents espouse either a perspectivistic or a categorical developmental orientation toward their children. The perspectivistic orientation views the child as an evolving individual whose immediate behavior is a function of individual characteristics and environmental factors and has been associated with healthy child adjustment (pg. 153). In contrast, the categorical orientation
21 views the child as a static entity and allows for little consideration of situational variables and individual differences (pg. 153). These developmental orientations, along with parental values, contribute to the attitudes that parents use to convey social skills to their child and result in either adaptive or maladaptive coping strategies. In relation to CD, the risk increases or decreases depending on the parents view and the environmental factors the child is exposed to through their development. Parental management of behavior Parental beliefs and attitudes are the foundation of how parents interact with their child with regard to management of behavior. Importantly, Gardner (1992) noted that many children with CD live in a family context where the reactions of parents are unpredictable. This unpredictability results from parents providing indiscriminate responses to their childs behavior. For example, it has been noted that children with CD are more likely than children without behavior disorders to receive indiscriminate responses from their parent(s), including aversive reactions to appropriate behavior (Gardner, 1992). Such findings suggest that it is through ineffective parenting practices that many family interactions occur in which coercive child behaviors are reinforced. Rutter (1985) emphasized the importance of consistent supervision/discipline and suggested four dimensions that are characteristic of families whose children develop behavioral problems. First and foremost is the lack of house rules to provide the children with clear expectations of what they may and may not do. Second is the lack of parental monitoring of the childs
22 behavior. This means that parents are not adequately informed about the childs acts or emotions, and hence, they cannot respond appropriately. Third is the lack of effective contingencies. This usually occurs when parents nag and shout but do not follow through with appropriate disciplinary action, or when they do not respond with an adequate differentiation between praise for prosocial activities and punishment for antisocial activities. The fourth and final dimension is a lack of techniques for dealing with family crises or problems. This results in conflicts that lead to tension and dispute but do not result in resolution. It is apparent that these parental behaviors can leave a child confused regarding limits and consequences, which may facilitate aversive behavior and result in the child being reinforced for escalating his or her aggressiveness. Gelfand et al. (1982) proposed that child-rearing practices consist of several different components, including control, affective-emotional, discipline, and psychological. These approaches can be considered extremes in relation to discipline styles that create a four-ended continuum. Although many families fall somewhere in between the extremes of the continuum, it appears that parenting styles that are located at the extremes influence the development of disruptive behavior disorders. For example, parents who use erratic control and are permissive are more likely to have aggressive and behaviorally disordered children. Likewise, children with a lenient, permissive mother and a rigid, restrictive father are typically aggressive and delinquent. While children learn how to approach and handle situations by observing parents discipline practices, they also learn behavior through modeling or
23 imitating their parents social skills. As explained by Robinson (1985), conduct problems are seen as stemming partly from a failure to develop the complex guidelines necessary for adequate social and ethical functioning in society. These guidelines are learned through direct and indirect experiences that are modeled in the home. Further, Robinson (1985) stated that, parents communicate with their children through their words, their actions, and their emotional ambience, and enhance communication when cues are congruent via all three channels (pg. 616). Therefore, when an adult verbalizes a particular principle but fails to respond behaviorally or emotionally when the child violates that standard, then child may not be able to apply those values. In addition, the failure of a parent to state the rules when rewarding or punishing may elicit a trial and error approach within the child. Gelfand et al. (1992) have noted that the act of being disciplined by a parent is possibly the most intense and vivid example of how to control another persons behavior. Therefore, parental responses that do not consider the childs source of motivation, feelings, and perceptions may contribute to the development of conduct problems by not providing the child with the ability to acquire models of moral behavior through processing the various experiences with the child (Robinson, 1985). In addition to the interaction style between parent and child, research suggests that there also are child factors that contribute to the development of CD and ODD. For example, Wooton, Frick, Shelton, and Silverthorn (1997) conducted a study examining the moderating role of callous-unemotional (CU) traits of children on ineffective parenting and childhood conduct problems.
24 Among these callous-unemotional traits, which the authors believe place a child at a high risk for showing antisocial behavior, are lack of empathy, lack of guilt, manipulativeness, and emotional constrictedness. These traits are consistent with low emotional reactivity, which is characterized physiologically by underreactivity in the autonomic nervous system. Behaviorally, this distinct temperament style is associated with lack of fearful inhibitions (Wooton et al., 1997), resulting in a child who is less responsive to cues of punishment, and hence, resistant toward parental and societal norms. Developmental research proposes that this temperamental style can be related to the development of CU traits in the following ways: (1) it could place a child at risk for missing some of the early precursors to empathic concern, (2) it could lead a child to be insensitive to the disciplinary efforts of parents and other socializing agents, and (3) it could create an interpersonal style in which the child expects instrumental gains from his or her aggressive actions, resulting in the inability to resolve interpersonal conflict appropriately (Frick, 2000). According to Wooton et al. (1997), children who have CD or ODD and who display CU traits develop problematic behavior differently than children who have CD or ODD without these traits. More specifically, the development of conduct problems among children with CU traits may be independent of parenting practices because their temperament style has made them unresponsive to socialization practices. On the other hand, children who do not display these traits will be highly susceptible to inadequacies in their rearing environment (Wooten et al. 1997). These authors therefore hypothesized that ineffective
25 parenting practices are more strongly associated with conduct problems in children without CU traits. To test their hyposthesis, Wooten et al. (1997) examined the predictive value of CU traits and ineffective parenting in the development of conduct problems by utilizing a three-step hierarchical multiple regression procedure. As hypothesized, their analyses revealed that children high on CU traits had high rates of conduct problems regardless of the quality of parenting. Further, it was found that ineffective parenting was positively correlated with conduct problems in children without CU traits. In a similar study, Frick, Christian, and Wooton (1999) examined age trends in relation to parenting practices and conduct problems. In response to (a) the lack of adequate methodology for examining parenting practices in early childhood through adulthood and (b) the typical focus on adolescents reports of these practices, Frick developed a multi-informant and multimethod system of assessment. This system considers the reports of both the parent and the child in an effort to examine parenting practices that are most closely associated with the development of ODD and CD. Participants were between the ages of 6 and 17 and were divided into three age groups: a young group (ages 6-8), a middle group (ages 9-12), and an adolescent group (ages 13-17). In an effort to measure parenting style, the Alabama Parenting Questionnaire (APQ) was administered. The NIMH Diagnostic Interview Schedule for Children (DISC) was utilized to assess the presence of each symptom of ODD and CD. The DISC is a structured
26 psychiatric interview that was administered to the childs parent (DISC-P), the childs teacher (DISC-T), and to the child (DISC-C) if he or she was older than 9. This procedure was used to ensure that symptoms were assessed through multiple informants at each age but limited it to those informants who seemed to be most valid at various age groups. The relationship between each of the five parenting constructs from the APQ and conduct problems in each age group was determined through a multiple regression analysis. The amount of variance in conduct problems as explained by the scores from each form of the APQ for a given construct was used as the estimate of the association between the parenting construct and conduct problems. These analyses were conducted twice in order to examine all the children with complete data across all forms of the APQ as well as to examine the sample while eliminating children who scored above the upper quartile on a callous-unemotional screening scale. As mentioned previously, (Wooton et al., 1997), children high on CU traits may develop conduct problems through a process that is independent of parenting practices. When they removed children who scored high on CU traits, Frick et al. (1999) found that the coefficient of determination (R 2 ) estimates between conduct problems and parenting were consistently higher (p 117). A hierarchical regression procedure also was conducted in which demographic variables were entered into the regression equation, which yielded results similar to those in the previous analysis.
27 The results of these analyses indicate that parent involvement, the use of positive discipline strategies, and level of supervision decrease as the child gets older (Frick et al., 1999). Furthermore, parental consistency in relation to discipline practices accounted for the largest amount of variance in conduct problems within the adolescent group. The authors, therefore, suggested that prevention and intervention programs for families of adolescents should include a component that focuses on increasing positive parental involvement with the adolescent while still respecting his or her increasing need for independence (p 121). Parental consistency was moderately predictive of behavior problems in the youngest age group, whereas corporal punishment was most strongly associated with CD in the middle age group (ages 9-12). Family Systems Theory Family systems theory incorporates all of the aforementioned variables into a comprehensive, theoretical way of examining the family as a whole. The premise of this theory is that people are part of a social context, and in order to develop an understanding of them, one must understand the family context, or system, as a whole (Fisher, 1996). A family system is not just a compilation of individual personalities and behaviors. Rather, it includes the complex interactions of all members and how they function together. Furthermore, the theory focuses on examining the events in the context in which they occurred rather than in isolation (referred to as a patterned non-linear approach to relationships) ( Mikesell, Lusterman, & McDaniel 1995).
28 The systemic principle that dominates family systems theory is the concept of interdependence (Robbins, Szapocznik, Alexander & Miller, 1998) which pertains to the complex connections and mutual influence of individual members within a family. The notion of boundaries captures the extent of interdependence within a family system through examining the familys functioning and relationship characteristics. There are several types of boundaries, such as the family subsystem boundary, which separates/connects systems within the family as a means to organize its many roles and functions. An example of this would be the parent-child subsystem boundary. This type of boundary is considered the most important subsystem boundary in family systems therapy (Robbins et al., 1998) because it denotes the extent to which a family can work as a unit to carry out roles and responsibilities. Interpersonal boundaries are a second type of boundary and represent the level of cohesion within the family unit. Some preliminary research has supported a family systems therapy approach to intervening with families who have a child with a DBD (Szapocznik et al., 1989, Robbins et al., 1998, and Kazdin, 1993). Specifically, Szapocznik et al. (1989) examined the effectiveness of a family therapy model in treating conduct-disordered children in comparison to individual treatment and no treatment. The results of this study revealed that family therapy had a positive effect on the reduction of problem behaviors and was considered more effective than individual therapy according to parents. The prevailing model that integrates
29 a holistic approach to family systems in examining the ways in which family members function is the Circumplex Model of Marital and Family Systems. The Circumplex Model of Marital and Family Systems Olson (1982) developed the concept of the Circumplex Model in an attempt to connect research, theory, and practice. The basic premise of the model is that family functioning can be determined by exploring a descriptive map of various types of couple and family relationships. A continuum containing three dimensions (i.e., family cohesion, flexibility, and communication) is utilized to assess the degree of balance the family possesses. According to Olson (1993), cohesion can be defined as the intensity of emotional closeness, or togetherness, that family members experience in their relationships with one another. The cohesion continuum has four distinct sections: disengaged, separated, connected, and enmeshed. Low levels of cohesion (whereby family members are highly independent and uninvolved with each other) characterize a family with a disengaged system. In contrast, an enmeshed family is overly dependent and extremely cohesive. Family functioning on either extreme is considered dysfunctional. A more functional family would fall somewhere away from the extremes of the continuum. For example, a separated family values time apart as important, yet they do spend some time together, make joint decisions, and provide each other with support. A connected family views time together as more important than time apart; however, they do have some separate friends and activities.
30 The second dimension, adaptability, refers to the amount of flexibility that family members permit in regard to rules/roles (Olson, 1983). A healthy balance is necessary for the family to be able to accommodate to the various developmental stages or daily stressors they may face. As with cohesion, the dimensions of adaptability are on a four-part continuum that is considered problematic at the extremes. The specific levels include: rigid, structured, flexible, and chaotic. Families that engage in relationships that are controlling with unchangeable rules and strictly defined roles are considered rigid; families that exhibit too much adaptability find themselves in a system with little control, instability, and insecurity described as chaotic. In contrast, a structured family engages in a democratic leadership style. The third dimension in the Circumplex Model, family communication, is considered a facilitating dimension. Adaptability and cohesion are reflected in the way the family communicates, meaning that there is a bi-directional relationship between family communication and family functioning. Communication within the family is assessed by examining the members as a group with regard to listening skills, speaking skills, self-disclosure, clarity, respect, and regard (Olson, 1993). It is family communication that enables families to address the levels of cohesion and adaptability and strive towards effective family functioning. It is important to note that there is no predetermined level of adaptability and cohesion that is considered ideal for relationships to be functional. Since families come from various cultural backgrounds, their values as well as their
31 comfort level regarding their intensity of interactions also vary. Therefore, it may be beneficial to identify the expectations of the individual members to determine their optimal level of adaptability and cohesion in addition to each members degree of family satisfaction. While there is no archetype of family functioning that ensures the ideal family situation, there are suggested points along the continuum that are associated with higher degrees of functionality. These balanced areas are the result of family members ability to experience unhealthy extremes of cohesion and adaptability and balance them. For example, a family that is balanced on cohesion is characterized as separated and connected, meaning that members are able to maintain equilibrium between being alone versus being together in a more functional way. As for flexibility, balanced systems (i.e., structured and flexible) have a more practical approach to change and stability (e.g., openness to change when necessary). Perceptions of Family Functioning Several studies have utilized the Circumplex Model of Marital and Family Systems to compare parents and adolescents perceptions of their own familys functioning. For example, Nollar and Callan (1986) conducted a study that assessed adolescents and their parents perceptions of adaptability and cohesion within intact, non-clinical families. The mother, father, and an adolescent from each participating family were given the Family Adaptability and Cohesion Evaluation Scales III (FACES-III, Olson, 1982) to determine their perceptions of how the family was currently and how they would ideally like the family to be.
32 The results revealed that adolescents viewed their families as less flexible and less cohesive and ideally wanted their families to be more flexible and less cohesive in comparison to their parents (Nollar & Callan, 1986). The trend seen in these findings corresponds to the theoretical suggestion that adolescents are in the process of developing autonomy and thus are separating themselves from their parents (Nollar & Callan, 1986). This may help to explain why adolescents were less satisfied with their family current functioning than were their parents. The Circumplex Model of Marital and Family Systems also has been utilized in research examining family functioning in families with a child diagnosed with CD. Slee (1996) conducted a comparative study between families with a daughter with Conduct Disorder and families with a typically developing daughter to investigate mothers perceptions of family climate. Participants consisted of 38 families, 19 of whom were in the clinic group and 19 of whom were in the control group. The mothers in each family were administered the Moos Family Environment Scale (FES) to obtain their perceptions of family climate as well as the climate of the family in which they were raised as children (family of origin). Further, a random sample of 18 families (nine each from the clinic and control groups) was videotaped in their homes for a 35-minute period, at which time the interactions between the mother and daughter were observed. Five activities were strategically selected as the agenda for the observational time so that there would be a wide variety of interactive opportunities.
33 When comparing the responses of the two groups of mothers on the FES, it was found that mothers of daughters with Conduct Disorder perceived their families as less cohesive, less encouraging of independence, and more controloriented. Results also suggested that they perceived their families as being less expressive of feelings, more expressive of open conflict, and less organized. Overall, the organization of the family as perceived by the mothers was lacking in structure and clarity regarding family rules and responsibilities. In looking at the intergenerational perspective, mothers of daughters with Conduct Disorder reported that their families of origin were less cohesive and less encouraging of independence than mothers of daughters without a diagnosed behavior disorder. Although it has been found that mothers of girls with Conduct Disorder differ substantially in their perceptions of family functioning as compared with a control group of mothers, to get a fuller picture of these families it is imperative to include other family members such as the child. Including other informants enables researchers to assess the differences between the responses of each individual. More importantly, it broadens the understanding of family dynamics and how they are maintained. For example, Noller, Seth-Smith, Bouma, and Schweitzer (1992) reported on two studies that compared the perceptions of clinic and non-clinic families and included both adolescents and their mothers. These studies were based on the generational stake hypothesis, which proposes that each generation views family functioning in terms of their own biases. Specifically, this theory suggests that adolescents hold more negative views of their families than do their parents because they are in the process of
34 establishing themselves as individuals and are separating from their families. Parents take on a more positive view since they are looking for validation of their efforts on behalf of their families (p 102). To test this hypothesis, the clinic sample in the first study consisted of 33 boys and girls who had been referred to a child guidance clinic. The subjects were matched for age, gender, family structure, and socio-economic status (SES) to comprise a non-clinic sample of 33 participants. Assignment to the nonclinic group also was based on a score that fell within the normal range on the Child Behavior Checklist (CBCL; Achenbach, 1979) as reported by the childs mother. In the second study, the clinic sample contained 30 pairs of adolescents and their mothers who attended a guidance clinic. These participants also were matched with 30 non-clinic adolescents and their mothers on variables including age, gender, family structure and SES. The independent variable for both studies was the clinical diagnosis (clinic and non-clinic) of the child, whereas the dependent variable included the perceptions of family functioning and self-concept. Measures utilized in this research study assessed the presence of behavior problems (CBCL: Achenbach, 1979), self-concept (the Self Description Scale: SDQ-III; Marsh, Parker & Barnes, 1985), and factors of family functioning (ICPS Family Functioning Scale ; Noller, 1988). A factor analysis was conducted on the latter scale, which yielded three factors of family functioning: (a) intimacy, (b) parenting style, and (c) conflict. A MANOVA was conducted for the first study in order to examine the differences between the clinic and control groups on family functioning variables
35 (i.e., intimacy, parenting style and conflict). No interaction effects were found, indicating that the two groups of adolescents perceived their families similarly. A canonical correlational analysis also was conducted to explore the relationship between the self-concept variables and the family functioning variables. Higher levels of intimacy and a more democratic parenting style were associated with higher scores on the self-concept dimensions, whereas higher levels of conflict were associated with lower scores on self-concept scales. In the second study, a MANOVA was utilized to assess the relationship between the perceptions of adolescents and their mothers with regard to family functioning. The results indicated that the adolescents viewed their families as less intimate, less democratic, and more conflictual than the mothers. Although no significant differences were found between the clinic and non-clinic groups of adolescents, there was a significant difference between the two groups of mothers, with the clinic mothers rating their families no more positively than their adolescent yet much more negatively than non-clinic mothers. This study illuminates the differences in perceptions among family members within the control group and the similarities between the mother and adolescent in the clinic group. More importantly, it emphasizes the significant difference in perception between the mothers of non-clinic adolescents and clinic mothers, as well as the lack of differences in perceptions between the clinic and non-clinic groups of adolescents. These findings support the notion that adolescents, clinic or nonclinic, go through a developmental process in which they view their family as more negative. Further, it suggests that there is a bi-directional relationship
36 between the mother and adolescent in the clinic family in that family members influence one another and often engage in a coercive cycle, eventually leading to similar, negative perceptions of their family. Given these findings, the generational stake hypothesis was supported for non-clinic families but not for clinic families. In other words, typically developing adolescents consistently rated their family as more negative than their mothers while clinic adolescents and their mothers rated their family as equally negative. The authors address this by suggesting that adolescents with behavior disorders eventually reach a level of aversive behavior at which point the caregiver can no longer maintain the positive view of the family and may give up their stake in the family. Pillay (1998) assessed the perceptions of adolescents with Conduct Disorder in comparison to their normally developing peers regarding levels of cohesion, adaptability, and family satisfaction. The adolescents with Conduct Disorder ranged in age from 13 to 16 and were recruited from the Child and Adolescent Unit at a hospital in South Africa. The control group was selected from various community groups such as schools, youth groups and sports clubs, and was matched on several variables (i.e., age, gender, living situation, and SES). The participants were administered the Family Adaptability and Cohesion Evaluation Scales (FACES III) and the Family Satisfaction Scale. These two self-report scales assess individuals perceptions of their families adaptability and cohesion and levels of satisfaction with their family functioning. A limitation
37 of this study noted by the author is that neither one of these instruments has been standardized or normed for the South African population (Pillay, 1998). Results of this research revealed that the adolescents with Conduct Disorder scored significantly lower than their control group peers on adaptability, cohesion, and satisfaction. In other words, the adolescents with Conduct Disorder perceived their families as significantly more rigid and disengaged and were generally less satisfied with their familys functioning. Although these results may provide insight on the family environment in which these adolescents were being reared and how it affects the development of CD, it may also be reflecting the negative, blaming behavior that is characteristic of CD (Pillay, 1998). Unfortunately, no other family members with whom the adolescents perceptions could be compared were included in this study. Summary The research on children identified with a DBD has supported the relationship between the etiology of CD and ODD and family variables. More specifically, the literature examining the characteristics of these families has found them to be less cohesive and less adaptive (Haddad et al., 1991; Noller et al. 1992; and Pillay, 1998) than families of typically developing children. In looking at the broader picture, there is a predisposition for children with early problems to continue on a chronically aggressive path (Lyons-Ruth, 1996). These children frequently exhibit numerous characteristics that distinguish them form children who do not have CD or ODD. The literature suggests that the chronically aggressive child (a) first displays symptoms in preschool years, (b) exhibits the
38 symptoms at a greater frequency and in more settings, (c) displays early hyperactive behavior, (d) and develops covert antisocial behaviors. Although these characteristics tend to be stable and appear to progress through ones lifetime, it is unlikely that all children who develop conduct problems do so as a result of the same causes. Further, it has been suggested that children who demonstrate CU traits often develop behavioral problems independent of family functioning. Although the literature provides considerable insight into the etiology of ODD and CD in addition to the characteristics of family functioning, there are many limitations of this research. First and foremost is the lack of a comprehensive research that includes preadolescent children. For example, several researchers have assessed perceptions of family functioning among adolescents in comparison to their primary caregivers, but little research has examined these issues with elementary-aged children. A second limitation of the research in this area is that no studies have specifically compared perceptions of family functioning among children with a disruptive behavior disorder and their parents. To date, studies have compared perceptions of family functioning among: (1) members of nonclinical, intact families, (2) adolescents diagnosed with ODD or CD and their non-clinical peers, (3) mothers of clinical and non-clinical children, and (4) clinical adolescents and their mothers to non-clinical adolescents and their mothers (Nollar & Callan, 1986; Pillay, 1998; Slee, 1996; Noller et al.). This study will expand the current literature base by comparing the perceptions of a preadolescent child diagnosed with ODD or CD in comparison to his or her primary caregiver.
39 Although developmental outcomes (i.e., antisocial behavior) are a result of a complex interplay of sociocultural, biological, and intrapsychic processes (Frick, 2000), the childs home and the environment the family creates are the basic forum within which development, healthy or maladaptive, occurs. Therefore, it is necessary to obtain a better understanding of perceived family functioning among families with a child diagnosed with a DBD. This research will provide additional information that will contribute to the understanding of family functioning and ODD and CDD in relation to how each member experiences their family.
40 Chapter III Method Participants The sample included 29 families consisting of a child who met the criteria for Oppositional Defiant Disorder (ODD) or Conduct Disorder (CD) and a primary caregiver. The children with ODD or CD were elementary school students between the ages of 7 and 12 who were receiving exceptional student education services through Emotionally Handicapped (EH) or Severely Emotionally Disturbed (SED) units from several schools within the Hillsborough and Pinellas County School Districts. All participants met criteria for ODD or CD as measured by the Disruptive Behavior Disorders (DBD) Checklist (Pelham, Gnagy, Greenslade & Milich, 1992). Since children with CD or ODD are commonly diagnosed with other disorders (i.e., Attention-Deficit/Hyperactivity Disorder), the participants with comorbid diagnoses were eligible to participate. The primary caregiver was identified as the adult in the household who spends the most time with the child. The criteria were the same if the parents were divorced, in which case the parent with custody was most likely the adult who spent the most time with the child.
41 Measures Family Demographic Information Sheet. Family demographics were obtained using a form developed by the principal investigator (see Appendix A). Data that were gathered included the following: (1) gender of primary caregiver and child with ODD or CD, (2) age of primary caregiver and child with ODD or CD (3) relationship of the primary caregiver to the child with ODD or CD, and (4) number of members living in the household. The primary caregiver was the source of information, which was gathered during a phone interview. Disruptive Behavior Disorders Checklist: Teacher Form (DBD-TF). The diagnosis of a Disruptive Behavior Disorder was determined through the Disruptive Behavior Disorders Checklist (DBD:TF; Pelham et al., 1992). This measure was downloaded from a website ( http://wings.buffalo.edu/adhd ) endorsed by the author (see Appendix B). The DBD-TF is a 45-item self-report rating scale designed to measure disruptive behavior symptoms based on the criteria specified in the Diagnostic and Statistical Manual of Behavior Disorders Third Edition Revised (DSM III R). Data collection on the DBD-TF began prior to the publication of DSM-IV, resulting in the focus on the earlier version of the DSM. According to the first author, items 10, 14, and 21 were originally included for the DSM-III-R and have been subsequently removed to align with the DSM-IV criteria for Disruptive Behavior Disorders. The four factors that are used to define disruptive behaviors include ADHD-Inattention, ADHDHyperactivity/Impulsivity, ODD, and CD. Respondents are required to indicate the severity of symptom occurrence as derived from a 4-point scale ranging from
42 0 (not at all) to 3 (very much). A response of dont know also is accepted if the rater does not have adequate knowledge of the behavior in question. Scoring of the scale is completed by adding up the teachers rating for each item that corresponds to the factor and dividing the total by the number of items the teacher completed on that factor. This average rating is then compared to the normative sample, also provided on the website. Eligibility for the current study was established based on a total score of 2 standard deviations above the mean for the childs age and sex, which is the cut-off suggested by the author. Pelham et. al (1992) obtained normative data for males on the DBD-TF through educational professionals who responded to an article geared toward recruiting participants. A total of 301 teachers completed the scales on 1,505 children ranging in age from 4 to 19 years from 48 states. Of the 1,505 completed ratings, 987 were of children from regular education mainstreamed classrooms, and 413 were obtained on children receiving special education services at some point in the day. Younger and older children were eventually excluded, resulting in a final sample of 931 boys between the ages of 5 and 14 from kindergarten through eighth grade. The authors realized through this norming procedure that teachers do not typically report characteristics of Conduct Disorder; rather the caregivers most often note these symptoms. In examining these phenomena further, the authors were unable to report internal consistency for the CD factor, resulting in final reliability coefficients of .96, .95 and .95 for ODD, ADHD-Inattention, and ADHD-Hyperactivity/Impulsivity respectfully. However, the DBD-TR still includes the CD factor and has been
43 utilized to explore the presence of such symptoms. It should also be noted that the authors are currently pursuing further research focusing on updating the psychometric properties and normative sample. Family Adaptability and Cohesion Evaluation Scales II (FACES-II ). Family functioning, in accordance with the Circumplex Model, was measured by the Family Adaptability and Cohesion Evaluation Scales II (FACES-II; Olsen, Portner, and Bell; 1982) (See Appendix C). This 30-item, self-report instrument was constructed to examine the frequency of occurrence of described situations on a five-point scale (1= almost never to 5=almost always) thus indicating how the informant sees his or her familys adaptability and cohesiveness. The scale is designed to measure the insiders view of family dynamics and focuses on all the family members currently living in the home. The FACES was developed in response to the need to validate the Circumplex Model. The scale initially consisted of 111 items focusing on three scales: Cohesion, Adaptability, and Social Desirability. Limitations in this first instrument lead to the development of the FACES II. More specifically, the initial FACES II was developed to address the following objectives: (1) to construct a shorter measure with more simple questions, enabling young children and those with a limited reading ability to use it, (2) to reduce the number of double negatives and provide a 5-point response scale, (3) to drop the Individual Autonomy Scale from cohesion, and (4) to develop a scale with two empirically reliable, valid, and independent dimensions (Olson, 1982).
44 The FACES-II was initially developed as the result of 464 adults responding to a 90-item measure. A factor analysis and reliability analysis were conducted, reducing the scale to 50 items. These 50 items were then administered in a national survey consisting of 2,412 individuals. The final measure was reduced to 30 items comprised of 16 cohesion items and 14 adaptability items. There are eight concepts related to the cohesion dimension including: emotional bonding, family boundaries, coalitions, time, space, friends, decision-making, and interest and recreation. An example of an item listed on the cohesion dimension is, Family members are supportive of each other during the difficult times. The total cohesion score for an individual should range between 16 and 80. In addition, there are six concepts pertaining to the adaptability dimension including: assertiveness, leadership, discipline, negotiation, roles and rules. An example of an adaptability item is as follows: It is hard to know what the rules are in our family. The range of scores for the adaptability scale is between 15 and 70. Reliability of the final version was established through internal consistency and test-retest reliability tests. Specifically, the total sample (N = 2,412) was divided into two equal subgroups, and the internal consistency of the items was then evaluated through Cronbachs alpha. The internal consistency coefficients for the first sample (N = 1,206) were r = .88 for the Cohesion subscale and r =.78 for the Adaptability subscale. The results of the reliability study for the second sample (N = 1,206) were similar, with coefficients of r = .86 for the Cohesion
45 subscale and r = .79 for the Adaptability subscale. The total internal consistency coefficient was r =.90. Reliability was determined through a test-retest study consisting of university and high school students (N = 124). The 50-item version was utilized and the time lapse between the two administrations was four to five weeks, resulting in a Pearson correlation of .83 for Cohesion and .80 for Adaptability. FACES Childrens Version Family functioning as perceived by the child with ODD or CD was assessed through the FACES II Childrens version (see Appendix D). Currently, the only childrens version available is in the form of a downward extension of the FACES-III. This measure was constructed by Dean Liskum (date unknown), who modified the adult version to yield a more developmentally appropriate evaluation for children and adolescents. The items match the content of the adult version, with the revision focusing on rewording the questions for easier readability and comprehension. However, the psychometric properties of this version are undetermined due to a lack of fieldtesting. Therefore, the principal investigator modified the FACES-II to create a childrens version for the participants in this study. The revised tool was analyzed by calculating Cronbachs alpha to determine its internal consistency reliability and is further discussed in the results section. This form of reliability assesses the extent to which the individual items that constitute a test correlate with one another or with the test total (Hatcher & Stepanski, 1999). Psychopathy Screening Device (PSD) Examining the presence of Callous-Unemotional (CU) traits in the identified child was accomplished by
46 administering the Psychopathy Screening Device (PSD; Frick & Hare, in press) (See Appendix E) to teachers. The PSD is an extension of the Psychopathy Checklist Revised (PCL-R; Hare, 1991), which has been utilized to identify psychopathic adults in forensic samples. Based on the promising findings in adults, the 20 items on the PCL-R each were revised into an analogous item that was more applicable to children (Frick & Eliis, 1999). Similar to the PCL-R, the PSD consists of a 3-point scale scored as 0 (Not at all true), 1(Sometimes true), or 2 (Definitely true). A factor analysis of the PSD was conducted using parent and teacher ratings of 95 clinic-referred children between the ages of 6 and 13. The findings were consistent with the adult forensic samples in that two factors emerged; one consisting of callus-unemotional traits and one consisting of problems of impulse control and conduct problems. Although this research suggests that the PSD is a promising measure for extending the multidimensional conceptualization of psychopathy to children and adolescents, there are limitations (Frick, Boden and Barry, 2000 ). First, the sample was small (N=95), primarily male (81%), and were all clinic-referred children. Second, validity studies, in addition to the factor analyses, included only children and adolescents who were adjudicated or clinic-referred. Third, there has been little focus on how the dimensions of psychopathy fit within the existing classification for childhood behavior disorders (Frick, Boden and Barry, 2000). Frick, Boden and Barry (2000) addressed these limitations by conducting a factor analysis of parent and teacher ratings on the PSD in two large samples of children. The community sample consisted of 1136 non-referred and non
47 adjudicated children in the 3 rd 4 th 6 th and 7 th grades, whereas the participants in the clinic referred sample were 160 children from an outpatient mental health clinic. To test the dimensionality of the PSD, an exploratory principal axis factor analysis was used for factor extraction, and an oblique (promax) rotation was used to examine factor loadings. A scree plot of eigenvalues listing each successive factor that was removed was inspected to determine the optimal number of factors. Then, prior to the factor rotation, the factor structure was refitted to optimize explained variance according the number of factors specified (Frick, Boden, & Barry, 2000). The scree plot revealed that either a two or three factor solution could be justified in the community and clinic samples, however the two-factor solution was more justifiable in the clinic-referred sample. The two factor solution led to one dimension consisting of narcissistic traits and impulsivity and the second dimension consisting of callous and unemotional traits. For the three-factor solution, the callous and unemotional dimension remained intact whereas the narcissism and impulsivity items divided into two separate factors. Next, the similarity of the factor structures for the two samples was tested by correlating the rotated factor loadings to yield an index of congruence. The results revealed a correlation of .90 across the two samples on the narcissism factor loadings, followed by correlations of .81 for the callous-unemotional factor and .68 for the impulsivity factor (Frick, Bodin, & Barry, 2000). In an effort to further determine the comparability between the two samples, a confirmatory factor analysis was conducted in the clinic-referred sample. This analysis was
48 conducted to examine how well the three factor structure from the community sample fit the data in the clinic sample. The fit indices revealed that the two factor structure was more appropriate for the clinic sample, and that adding the third factor to the model did not lead to a significant increase in the models fit (Frick, Boden and Barry, 2000). Internal consistency analyses were conducted to examine whether there was further support for distinguishing between the impulsivity and narcissism items. Although the separation of the items was previously justified through a factor analysis in the community sample, Frick, Bodin, and Barry (2000) suggested that this type of external validation is important in choosing between different factor structures. Thus, three subscales of the PSD were formed. The Narcissism subscale consisted of 7 items and had a coefficient alpha of .83 in the community sample and .85 in the clinic sample. The 5-item Impulsivity subscale had alphas of .74 and .64 in the community and clinic samples, respectively. Lastly, the 6-item Callous Unemotional subscale had internal consistency coefficients of .76 in the community sample and .65 in the clinic sample. All three subscales were highly intercorrelated, with the Narcissism and Impulsivity subscales showing the highest correlations in both samples. Analyses also were conducted to assess the associations between the PSD subscales and the DSM-IV diagnoses. Overall, there was a high association (r =.50, p <.001 to r =.74, p <.001) with regard to the DSM-IV definitions of ADHD, ODD, and CD. More specifically, the Impulsivity subscale was more strongly associated with the criteria for ADHD (r=.45, p <.001), and the
49 Narcissism subscale seemed to be more strongly associated with the ODD criteria (r =.51, p <.001). Although further study is needed to establish its validity, the previous analysis supports the notion of distinguishing between the Narcissism and Impulsivity dimensions (Frick, Bowen, & Barry, 2000). Procedure The principal investigator obtained district IRB for approval to conduct the study within Hillsborough and Pinellas Counties. The schools were initially selected based on the presence of Emotionally Handicapped (EH) and Severely Emotionally Disturbed (SED) units. A total of four sites were recommended based on the presence of an exceptional education services program and selfcontained enrollment. The principal of each individual school was then contacted and permission was obtained for the primary investigator to use their site to recruit participants. Additionally, the principal investigator met with all teachers who taught EH or SED classes at the schools whose principals agreed to participate to explain the study as well as their role. All teachers (N=8) expressed their willingness to participate and consequently assisted in the data collection. The examiner sent home a letter explaining the study in addition to a parent consent form (See Appendices F & G), requesting that the consent form be sent back to school with their child for the teacher to collect. For those students who did not returned the forms within a week, a second letter and consent form was sent home. Anonymity and confidentiality was ensured for those students and their families who opted to participate. Specifically, each
50 participating student was matched with an identification number, which was used throughout the study. The name of the student and their family members was kept separate from the study materials in a filing cabinet and was purged by the principal investigator upon completion of the study. Pilot The principal investigator initially attempted to include family units consisting of a primary caregiver, the child with a DBD, and the childs sibling. This proposed procedure required the caregiver to complete all three rating scales for each child in a 30 minute phone interview. In addition, the identified child and sibling were asked to complete a questionnaire in a group setting at their respective schools. Several barriers arose throughout the implementation of this procedure making data collection challenging. First and foremost, many families did not meet the criteria for having two children within the specified age ranges. For those families who did meet the criteria, the interview process via phone was more lengthy than anticipated and the questions on each of the measures were difficult for parents to comprehend. This was particularly evidenced by the frequency of clarification questions asked by the caregivers. A second critical barrier was noted during the group administration of the FACES for the children. Specifically, the children experienced difficulty in understanding the Likert-type response system (1 = Almost Never to 5 = Almost Always) and often looked on to a peers questionnaire for guidance. It was believed that the data collected throughout this initial attempt was not reliable and therefore revisions were made to the procedure and to the childrens questionnaire. These details are discussed in detail in the following section.
51 Study A second attempt at data collection began, once again, with the recruitment of participants in several Hillsborough and Pinellas County Schools. The aforementioned steps described in the section prior to the pilot were implemented with modifications made to the roles of the participants. More specifically, this revised procedure utilized the classroom teacher as a means to obtain information on the severity of the childs behavior as well as Callous Unemotional traits, thus requiring the parent only to complete the family functioning questionnaire. An additional modification included the omission of the sibling, leaving the primary caregiver and child with a DBD as the defined family unit. This enabled the inclusion of more families, ultimately lending to a greater sample size. Additionally, changes were made to the FACES-Childrens version, which included rewording the possible responses as well as inserting an accompanying pictorial representation for each. This was achieved by illustrating a stick with beads denoting the amount associated with the childs perception (i.e., zero beads = no, never, one bead = no, two beads = sometimes yes/sometimes no, three beads = yes, and four beads = yes, always). As mentioned previously, a parent information letter and a request for consent were sent home in an effort to solicit participants. A total of 85 letters were sent home with children across 4 schools, of which 33 were returned. When a consent form was returned, the principal investigator provided the classroom teacher with the DBD-TF and PSD to complete on the individual child. The former scale was scored to determine if the child met eligibility for the study. Based on the cut off criteria of 2 standard
52 deviations above the normative sample, it was found that all children met eligibility to participate in the current study. The primary investigator then called the parents of the children who met eligibility to thank them for agreeing to participate and either scheduled a phone interview, or conducted the interview at the time of the initial call, according to parent preference. All interviews were conducted by the principal investigator and began with the demographic information sheet, followed by the 30-item Family Adaptability and Cohesion Evaluation Scale (FACES; Olson, 1982). The children who met the predetermined level on the DBD were given the FACES-II Childrens Version in an individual or small group format. All questionnaires were administered by the principal investigator at a location in the school building selected by the principal or classroom teacher. Examples of such locations include the media center, the school psychologists office, and teacher planning rooms. Child assent was obtained prior to the administration of the questionnaire with emphasis on confidentiality and voluntary participation (see Appendix H). A set of standard instructions were read to the individual and/or group orally (see Appendix I), followed by each item of the FACES-II Childrens Version to control for levels of reading ability and fluency. The sessions varied from 10 to 20 minutes. Interviews were not completed with four of these families due to either disconnected phones or child absence from school during the child interview dates. All participants (e.g., primary caregiver, identified students, teachers) were given the opportunity to ask questions about the study upon completion of all
53 items, at which time the interviewer debriefed the participants (see Appendix J). Further, all participating children received an edible treat for completing the items, and all families were provided with a $10 money order and a letter thanking them for their time and participation (see Appendix K). Research Design and Statistical Analysis This study used a cross-sectional, descriptive research design. This nonexperimental design examines data that are collected from participants during a single, relatively brief time period. Further, the data are directly applied to each case within that single time period, and comparisons are made across the variables of interest (i.e, the predictor variable) (Johnson, 2001), making the design the most appropriate for this study. Prior to analysis, the data were screened for accuracy. Specifically, the data were entered into the SPSS system, and then reentered, for purposes of data verification and detecting inconsistent entries. Demographic characteristics of the sample were calculated and basic descriptive statistics, such as the mean and standard deviation, were collected to provide a description of the sample characteristics. The first analysis examined the internal consistency of the various measures used, comparing the coefficients to those reported in the research. Next, the data were analyzed to determine how each individual family member (primary caregiver and child with a DBD) perceived their familys adaptability and cohesion. Marginal means were calculated for the two groups to determine if there were initial group differences and/or possible interactions.
54 The data were then analyzed using a repeated measures ANOVA. This analysis examined the differences in perception between the two family members, addressing the research question focused on exploring the presence of a difference between each members perceptions. Since in this type of analysis the variability due to individual differences is removed from the error term and individual differences are the major reason for error variance, this design is considered more powerful than completely randomized designs. Finally, the relationship between CU traits and child perceptions of family functioning was examined, addressing the exploratory questions of the study. The sample of children with a DBD were placed on a continuum based on their ratings of CU traits ranging from low to high scores and analyzed by means of a Pearson-Product Moment Correlation in an effort to examine the strength of the relationship between a childs perception of family functioning and his or her CU traits.
55 Chapter IV Results This chapter provides a description of the results derived from the statistical analyses used to address the research and exploratory questions and is discussed in four sections. First, the participants demographic characteristics are presented followed by the internal consistency reliability estimates of the measures used. Next, marginal means are provided in addition to a repeated measures ANOVA addressing the degree to which family members differ in their perceptions of family functioning. Finally, a correlation matrix denoting the relationship between childs perception of family functioning and presence of CU traits is provided. Demographic Data A total of 29 family units (i.e., caregiver and child) participated in the current study, resulting in 58 individual participants. Demographic information was provided by the primary caregiver in each family through their response to several questions pertaining to the following: age of child, grade level, ethnicity, caregiver age, relationship to child, and highest level of education attained. The majority of the children in this study were male (75.9%) and non-white (62%), with their mothers most often identified as the primary caregiver (79.3%). Seventy-six percent of the caregivers were below the age of 50, with 58.6%
56 holding a high school diploma or GED equivalent. These findings are summarized in greater detail in Table 1. Table 1. Participants Demographic Characteristics Male Female Demographic Characteristics (Child) N % N % Age by Ethnicity White Seven 2 6.9 0 0 Eight 3 10.3 0 0 Nine 3 10.3 0 0 Ten 0 0 0 0 Eleven 2 6.9 1 3.4 Twelve 1 3.4 0 0 Non-White Seven 1 3.4 0 0 Eight 1 3.4 1 3.4 Nine 1 3.4 2 6.9 Ten 4 13.8 1 3.4 Eleven 3 10.3 1 3.4 Twelve 1 3.4 1 3.4 Grade Level Second 7 24.1 0 0 Third 4 13.8 2 6.9 Fourth 3 10.3 2 6.9 Fifth 8 27.6 2 6.9 Seventh 0 0 1 3.4
57 Table 1. Participants Demographic Characteristics (cont.) Demographic Characteristics (Caregiver) N % Caregiver Relationship to Child Mother 23 79.3 Father 3 10.3 Grandmother 2 6.9 Foster Parent 1 3.4 Age of Caregiver 20-29 4 13.8 30-39 12 41.4 40-49 6 20.7 50-59 2 6.9 60-69 1 3.4 70-79 1 3.4 No Response 3 10.3 Education of Caregiver Did not finish high school 4 13.8 High School Diploma 17 58.6 Some College 3 10.3 Four Year College Degree 4 13.8 Completed Graduate School 1 3.4 Internal Consistency Reliability Estimates To obtain a measure of internal consistency for the instruments used in the current study, Cronbachs alpha was calculated for various subgroups as well as for the total sample on each scale. This procedure was conducted for the FACES-II, the FACES-Childrens Version, and the PSD. The analysis of the FACES-II yielded coefficients of .81 and .59 for the Cohesion and Adaptability scales respectively, with a total internal consistency reliability alpha of .82. These results are not completely commensurate with the coefficients reported by the author of this scale. Specifically, the author reported coefficients ranging from .86 to .88 for Cohesion and .78 to .79 for Adaptability, with an overall
58 reliability coefficient of .90. The internal consistency of the childrens version, as calculated using the current sample, is r=.67 for the Cohesion scale and r =.78 for the Adaptability scale. The overall coefficient is r=.84. As stated previously, reliability estimates were not conducted using a normative sample for the childrens version and therefore, a comparison cannot be made. Lastly, Cronbachs alpha was calculated for the PSD, resulting in a coefficient of r =.35 for CU traits. This is in comparison to the authors finding of .65 and .76 for clinic and community samples (n=1296), respectively. When interpreting these results, it is important to note the small sample size in the current study and differences between the samples, namely the demographic and educational status of the children. More specifically, the normative sample was predominantly Caucasian (77%) and was taught within a regular education setting (79%). This is vastly different from the current study in which Caucasian was the minority race (38%) and 100% of the children were served within a special education environment. These group differences, accompanied by a small sample size (29 completed rating scales with 6 items tapping into the construct of CU traits) contributes to the difficultly with which we can be confident that we are tapping into the construct of CU traits. Another issue that compromises the level of confidence is measurement error. That is, there are factors that may be contributing to the obtained scores on the PSD, therefore hindering the ability to acquire a true measure of CU traits. Examples of possible factors that may have influenced measurement error on the PSD include: 1) the teacher not understanding the question on the scale and providing a rating
59 anyway, 2) the teacher not truly knowing the answer to the question and guessing, and 3) the teacher rating the child based on how he/she thought was expected. See Table 2 for a summary of Cronbachs alpha coefficients for the aforementioned scales. Table 2. Cronbachs Alpha for the Rating Scales Measure Cronbachs Alpha Family Adaptability and Cohesion Evaluation Scale II Total .82 Adaptability .59 Cohesion .81 Family Adaptability and Cohesion Evaluation Scale Childrens Version Total .84 Adaptability .78 Cohesion .67 Psychopathy Screening Device Teacher Report Callus Unemotional Traits .35 Marginal Means and Repeated Measures ANOVA Mean scores on the FACES-II were calculated to examine potential differences between caregivers and children with DBD. Overall, children in this study rated their families as less cohesive and less adaptive than did their caregivers, with caregivers indicating that their family is flexibly separated. This suggests that caregivers view their families as balanced. Interestingly, the children, on average, also reported their family to be balanced, with a profile of structurally connected. The interpretation of the childrens scores should be conducted with caution, however, as normative data have not been obtained for
60 preadolescents and therefore the mean score for this group cannot be validly classified through the profile. The specific means and standard deviations for the sample are shown in Table 3 while the means and standard deviations for the specific breakdown of the categories are reported in Table 4. According to these results, 62% of the children in the study consider their family balanced in comparison to 69% of the caregivers. In contrast, 10% of caregivers perceived their family as extreme while 21% viewed the family as being in the mid-range. Childrens view of their familys family functioning was comprised of 14% and 24% in the extreme and mid-range respectively. Table 3. Descriptive Statistics for Family Functioning Measure N M SD FACES Cohesion (Child) 29 57.34 6.991 FACES Adaptability (Child) 29 44.34 9.213 FACES Cohesion (Caregiver) 29 63.55 9.199 FACES Adaptability (Caregiver) 29 50.48 6.473 Table 4. Descriptive Statistics for the FACES Categories Category Caregiver n(%) Child n(%) Extreme 3 (10.3) 4 (13.8) Chaotically Disengaged Chaotically Enmeshed 1 (3.4) 2 (6.9) Rigidly Disengaged 2 (6.9) 2 (6.9) Rigidly Enmeshed Mid-Range 6 (20.7) 7 (24.1) Chaotically Separated 1 (3.4) Chaotically Connected 3 (10.3) 1 (3.4) Flexibly Disengaged 1 (3.4) Flexibly Enmeshed 1 (3.4) 1 (3.4)
61 Category Caregiver n(%) Child n(%) Structurally Disengaged 1 (3.4) Structurally Enmeshed Rigidly Separated 4 (13.8) Rigidly Connected Balanced 20 (68.9) 18 (62.1) Flexibly Separated 3 (10) 3 (10) Flexibly Connected 6 (20.7) 6 (20.7) Structurally Separated 9 (31.0) 6 (20.1) Structurally Connected 2 (6.9) 3 (13.8) Mean scores also were calculated for the individual items to provide a greater level of specificity regarding differences in perceptions among the groups of parents and children. All participants responded to the items based on a five point scale such that 1=almost never, 2=once in a while, 3=sometimes, 4=frequently, and 5=almost always. Notably, several of the items in the measure are negatively worded. therefore, higher ratings on these items should be interpreted as the informant having a more negative view of the family. With regard to the Cohesion scale, the most notable differences between informants were related to the physical activities the family engages in, with children reporting a higher occurrence of separation. More specifically, children perceived family members as going their separate way more frequently (M=3.69, SD=1.04) than did the caregivers (M=2.10, SD=1.01) and gathering in the same room as less often (M=2.93, SD=1.16) than caregivers noted (M=4.34, SD=1.04). Other notable differences between the two groups of informants were found on items such as, we have difficulty thinking of things to do as a family and family members feel closer to people outside the family than to other family members.
62 The opposite trend in responses was observed on items pertaining to acceptance of each others friends and spending free time with family members. On these items, children indicated that this is the case more often than the caregivers noted. Both groups were more in agreement on items such as, family members go along with what the family decides to do and family members feel very close to each other. The means and standard deviations of all items on the Cohesion scale are reported in Table 5.
63 Table 5. Descriptive Statistics for the Cohesion Subscale of the FACES. Higher ratings on these items are interpreted as less cohesive due to the negative wording of the item. Cohesion Scale Items Child M (SD) Parent M (SD) Differ 9.* In our family, everyone goes his/her own way. 3.69 (1.04) 2.10 (1.01) 1.59 5. Our family gathers together in the same room. 2.93 (1.16) 4.34 (1.04) -1.41 15.* We have difficulty thinking of things to do as a family. 2.76 (1.41) 1.86 (1.25) 0.90 19.* Family members feel closer to people outside the family than to other family members. 2.41 (1.24) 1.66 (1.05) 0.75 29.* Family members pair up rather than do things as a total family. 2.66 (1.08) 1.93 (1.28) 0.73 25.* Family members avoid each other at home. 2.17 (1.07) 1.69 (1.07) 0.48 3.* It is easier to discuss problems with people outside the family than with other family members. 3.24 (1.27) 2.86 (1.25) 0.38 1. Family members are supportive of each other during difficult times. 3.93 (.923) 4.24 (1.15) -0.31 11. Family members know each others close friends. 3.83 (1.10) 4.14 (1.27) -0.31 30. Family members share interests and hobbies with each other. 3.93 (.704) 4.14 (1.06) -0.21 7. Our family does things together. 4.10 (.939) 4.21 (.902) -0.11 13. Family members consult other family members on personal decisions. 3.52 (1.21) 3.62 (1.21) -0.10 21. Family members go along with what the family decides to do. 3.83 (1.04) 3.79 (1.01) 0.04 17. Family members feel very close to each other. 4.55 (.783) 4.48 (.986) 0.07 27. We approve of each others friends. 3.72 (.922) 3.38 (.979) 0.34 23. Family members like to spend their free time with each other. 3.90 (1.05) 3.45 (.985) 0.45
64 Individual items on the Adaptability subscale also were examined, illustrating that the groups of caregivers and children differ the most in areas regarding division of household responsibilities, expression of opinion, and discipline. More specifically, the children rated their family as less adaptive in these areas as compared to primary caregivers. Interestingly, the opposite trend was revealed on issues addressing problem solving. Here, the children reported the family as more adaptive than did their caregivers. This was particularly evident on items such as, In solving problems, the childrens suggestions are followed and Our family tries new ways of dealing with problems. The item with the least amount of variability between informants (X 1 X 2 =0.07) was observed on item 26 stating, When problems arise, we compromise. The means and standard deviations of all items on the Adaptability scale are reported in Table 6. A more in-depth analysis of these differences was then conducted utilizing a repeated measures analysis of variance (ANOVA). This analysis showed significant differences in perceptions of family functioning for both Cohesion, F (1,57) = 31.236; p < .0001 and Adaptability, F (1,57) = 24.996; p < .0001. Children with elevated scores on the DBD rating scale consistently produced lower scores on the Cohesion and Adaptability scales in comparison to their primary caregiver. Figure 1 provides a graphic display of the FACES-II mean scores obtained by each informant on the scales. Specifically, children obtained a mean score of 57.34 ( SD = 6.930) compared to a mean score of 63.55 ( SD = 9.118) for caregivers on the Cohesion scale. Similarly, the children obtained a
65 Table 6. Descriptive Statistics for the Adaptability Subscale of the FACES. Higher ratings on these items are interpreted as less adaptive due to the negative wording of the item. Adaptability Scale Items Child M (SD) Parent M (SD) Difference 22. In our family, everyone shares responsibilities. 3.03 (1.40) 4.31 (1.07) -1.28 2. In our family, it is easy for everyone to express his/her opinion. 3.03 (1.24) 4.21 (.978) -1.18 18. Discipline is fair in our family. 3.38 (1.35) 4.38 (1.08) -1.00 14. Family members say what they want. 3.17 (1.63) 4.07 (.961) -0.90 28.* Family members are afraid to say what is on their minds. 2.69 (1.31) 1.79 (1.24) 0.90 24.* It is difficult to get a rule changed in our family. 2.90 (1.21) 2.17 (1.28) 0.73 6. Children have a say in their discipline. 2.07 (1.53) 2.62 (1.21) -0.55 4. Each family member has input regarding major family decisions. 3.52 (1.33) 3.76 (1.15) -0.51 10. We shift household responsibilities from person to person. 2.86 (1.33) 3.14 (1.64) -0.28 8. Family members discuss problems and feel good about the solutions. 3.62 (1.32) 3.83 (1.31) -0.21 26. When problems arise, we compromise. 3.90 (1.15) 3.83 (1.04) 0.07 20. Our family tries new ways of dealing with problems. 3.72 (.882) 3.38 (1.02) 0.34 12. It is hard to know what the rules are in our family. 2.34 (1.11) 1.86 (1.1) 0.48 16. In solving problems, the childrens suggestions are followed. 3.59 (1.24) 3.07 (.923) 0.52
66 mean score of 44.24 ( SD = 9.132) on the Adaptability scale, whereas caregivers obtained a mean score of 50.48 ( SD = 6.416). While it is natural to obtain low agreement rates between informants, it is not possible with the current data to determine the meaningfulness of the differences. This is primarily due to the FACES-Childrens Version not being validated, resulting in the reliance on adolescent norms. Correlation Matrix In an effort to ascertain the relationships between caregiver and child perceptions of family functioning, the scores for each informant were entered to form a scatterplot, thus denoting the magnitude of correspondence. A pictorial 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Cohesion Adaptability Subscale FACES Mean Score Child Caregiver Figure 1. Family Adaptability and Cohesion Evaluation Scale mean scores for child and caregiver on the Cohesion and Adaptability scale.
67 representation is presented in Figures 2 and 3. According to the scatterplots, there is minimal correspondence between informants on both constructs, suggesting that caregivers and children in the current study view their familys functioning differently. Notably, the outlying dyad for the Cohesion scale is different from the outlying dyad for the Adaptability scale. Figure 2. Scatterplot of Parent and Child Responses on the Cohesion Subscale of the Family Adaptability and Cohesion Evaluation Scales. Faces Cohesion-Parent 80 70 60 50 40 Faces Cohesion-Child 80 70 60 50 40 30
68 Figure 3. Scatterplot of Parent and Child Responses on the Adaptability Subscale of the Family Adaptability and Cohesion Evaluation Scales. This relationship was further analyzed through the generation of a correlation matrix shown in Table 7. Included in this matrix are the adaptability and cohesion scores for both caregivers and children, revealing Pearson r correlation values and level of significance associated with each analysis. Consistent with the scatterplots, the results gleaned from the correlation analysis reveal only a mild relationship between the informants (i.e., caregiver versus child) for each construct. Of particular interest is the stronger correlation within informants, that is, child responses on cohesion and adaptability (r=.669) as well Faces Adaptability-Parent 70 60 50 40 30 Faces Adaptability-Child 70 60 50 40 30 20
69 as caregiver responses on cohesion and adaptability (r=.495). As stated previously, this phenomenon implies that the caregivers and children in the current study rated their perceptions of their family functioning differently. An additional correlation analysis was conducted excluding the two outlying dyads, resulting in a reduced correlation. Table 7. Correlations Across Raters and Responses on the FACES FACES Cohesion (Child) FACES Adaptability (Child) FACES Cohesion (Caregiver) FACES Adaptability (Caregiver) FACES Cohesion (Child) 1.00 FACES Adaptability (Child) .669** 1.00 FACES Cohesion (Caregiver) .472** .287* 1.00 FACES Adaptability (Caregiver) .126 .317* .495** 1.00 ** p <0.01. p <.05. The exploratory questions regarding the relationship between CU traits among children with a DBD and their perceptions of family functioning also were addressed through Pearson Product Moment Correlations. The matrix showing this correlation analysis is presented in Table 8. It reveals a low correlation between CU traits and childrens perceptions of their familys adaptability ( r = .164, p = .397) as well as between CU traits and childrens perceptions of their familys cohesion r = .023, p = .907). Notably, however, the mean score of
70 childrens CU traits, as measured by the PSD, was 4.34, which falls below the 75 th percentile range of 5.0 according to the published norms. Additionally, it is important to note that the range of teacher responses was between the minimum and maximum scores of 1 to 8 on the CU domain of the PSD. These analyses indicate that children in the current study do not rate high on CU traits nor are the variables (CU traits and childrens perception of family functioning) related. Table 8. Correlation Between CU Traits and Childs Perception of Family Functioning PSD CU Traits FACES Cohesion (Child) FACES Adaptability (Child) PSD CU Traits 1.00 FACES Cohesion (Child) .023 1.00 FACES Adaptability (Child) .164 .669** 1.00 ** p< 0.01
71 Chapter V Discussion The current study explored perceptions of family functioning among children with a disruptive behavior disorder and their primary caregivers. Specifically, the sample included 29 family units consisting of a primary caregiver and a child with a DBD being served in an Emotionally Handicapped or Severely Emotionally Disturbed unit within the Hillsborough or Pinellas County school system. All children were in self-contained classes. Analyses examined the differences between participants perceptions of their familys functioning along the dimensions of cohesion and adaptability. Additionally, the study investigated the relationship between childrens perceptions of their families functioning and their teacher-reported level of callous-unemotional traits. The current chapter will discuss the results of these data analyses as well as implications for practice. Limitations of the current study also will be discussed, as will directions for future research. Differences in Caregiver and Child Perception of their Familys Family Functioning Descriptive data from the current study indicate that caregivers consistently produced higher scores on the Cohesion and Adaptability scales of the FACES than did preadolescent children who experience behavior difficulties.
72 The mean score on cohesion was 63.55 for caregivers as compared to 57.34 for children, revealing a difference of 6.21. Similarly, the difference between the mean caregiver rating for adaptability (50.48) and the mean adaptability rating for the children (44.34) is 6.14. These scores were significantly different from each other and are consistent with previous research showing that adolescents with a DBD typically rate their families as less cohesive and less adaptable than do their parents (Nollar & Callan, 1986, Noller, Seth-Smith, Bouma, and Schweitzer, 1992 & Pillay, 1998). Despite the noted differences, the categorical profile for the groups of caregivers and children fell in the Flexibly Separated and Structurally Connected ranges respectively, both of which are considered balanced. A greater level of specificity was obtained by examining the overall trend of perceptions as provided by the profile. In general, a majority of children (69%) and caregivers (62%) rated their family within the balanced range. It is important to note, however, that the childs profile is based on adolescent norms because norms for younger children are not yet available. Given this, the reliability and validity of the results are compromised. Even with this limitation, it can be determined that preadolescents and caregivers in the current study view their families differently. Validation of the childrens version is needed to ascertain the meaning of this difference in a practical sense. While there is a strong suggestion that preadolescents view their family as less cohesive and adaptive, the usefulness of such data has not yet been determined, by this study, supporting the need for future research.
73 Notably, the fact that adolescents with a DBD rate their families as less cohesive and adaptive than their parents do has been hypothesized to be due to the tendency for adolescents to respond with resistance and develop an overt mindset or attitude that varies from their caregivers or other adults. This study has shown, however, that elementary-aged children with a DBD also view their families differently than their parents. There are several possible explanations for this. The first possible explanation is that caregivers tend to portray a positive view regarding their family as a means to validate their parenting efforts. This phenomenon has been coined the generational stake hypothesis and serves as a way to explain the differences in perceptions of family functioning as rated by caregivers and typically developing adolescents (Noller, et. al, 1992). However, these authors found that the hypothesis regarding caregivers positive perception does not hold true for families of adolescents with behavior disorders. According to Noller et. al (1992), this can be explained by parents relinquishing their stake in the family, and their effort in portraying a positive view, in light of their childs increased levels of behavioral issues. In other words, the parents essentially give up on painting a rosy view of their family for outsiders because the severity of the childs behavior has increased. There is no research to date that applies this hypothesis to other age groups, such as preadolescents; however, the current study contributes to this line of research by generating additional hypotheses. For example, the parents in the current study consistently rated their family as more cohesive and more adaptive than did the preadolescents
74 with a behavior disorder. Therefore, it may be the case that the children have not reached the specific levels of aversive behavior discussed previously, resulting in the preservation of the optimistic outlook. There are several other lenses through which the findings of this study might be viewed. First, the discrepancies in perceptions of family functioning between child and caregiver may be a function of the childs inaccurate view of the family, relating to Dodges (1993) theory that these children interpret more hostile cues or lack the problem solving skills necessary for successful interactions. Specifically, if the child perceives their family differently, that is, more negatively, they may be misinterpreting the actions of other family members and be reacting based on their false perceptions. This could then lead into a coercive pattern of interactions with caregivers, thus allowing opportunities for the child to practice and consequently strengthening the aversive behaviors. Second, these children may have had a difficult temperament present since birth and therefore have responded to their environment differently in comparison to children who maintain an easy temperament. According to Thomas and Chess (1984), there are several traits that encompass a difficult temperament including high activity level, unpredictable, initial withdrawal, poor adaptability, intense, and negative. It has been noted that all of these traits combined describe approximately 10% of children (Thomas & Chess, 1984) and that a child demonstrating a combination of these traits requires more than good enough parenting. The premise behind good enough parenting is that easytempered children are not difficult to parent due to the positive and adaptable
75 traits of the child, resulting in the success of basic parenting techniques. As such, a child with a difficult temperament requires more skill on the part of the caregiver. It is the goodness of fit between the childs temperament and the parenting style or the environment that appears to be a strong predictor of a developmental trajectory significant for behavior problems. More specifically, Thomas and Chess (1984) have suggested that children with a difficult temperament are at a great risk for developing behavior problems. Given these findings and interpretations, it is important to validate the childrens version of the faces as a means to explore the differences in perceptions between typically developing preadolescents and their caregivers to ascertain whether or not the differences found in the current study are specific to children with a DBD. Additionally, it appears critical to include the perceptions of various family members, such as siblings to examine the differences as well as to obtain additional views of the family. This is discussed further in the future directions section of this chapter. Childs Perception of Family Functioning and Callous-Unemotional Traits This exploratory question investigated the relationship between the perceptions of family functioning as noted by the child with a disruptive behavior disorder and the presence of CU traits as reported by the childs teacher. The results of the correlation indicate that there is no statistically significant relationship between these two constructs. No research has been conducted on this relationship previously; however, the nature of CU traits encompasses lack of empathy and general lack of emotionality (Wooten, et al., 1997) feeding into
76 the hypothesis that children high on these traits would not view their families as cohesive and adaptable. There are several possibilities to explain the non-significant findings, including the poor internal reliability estimates found in the current study and the restriction of range on the CU traits variable as noted previously. For example, the participating teachers in the current study did not rate the children high on these traits, therefore limiting the validity of the analysis. In other words, there was not an elevated level of CU traits in the children, which may make it difficult to obtain significant findings when comparing this variable to childs perception of family functioning. It also is important to note that the perceptions of the caregiver may differ regarding the presence of CU traits and could have an effect on the relationship between the two variables. Another possible explanation for this non-significant finding is that the children in the current study truly do not demonstrate CU traits, rather they have developed behavioral problems due to inadequate parenting or other contributing environmental factors. This would lend support to the literature regarding CU traits as a mediating factor in the manifestation of conduct problems (Wooton, et. al, 1997) in that children without these traits have been found to be more susceptible to ineffective parenting techniques while children with CU traits develop behavioral difficulties regardless of parenting efforts. Furthermore, the recent development in the diagnosis of CD has focused on early onset of the disorder, where approximately 25% of the children have the characteristics of CU traits. Given the severity of this disorder, it would be atypical to find these
77 children in a typical classroom setting, which provides additional rationale for why children in the current study did not rate high on these traits. These findings also suggest that that parenting plays an integral part in the maladaptive behavioral patterns for this particular group. Given that the children in the current sample did not demonstrate CU traits, further exploration into childrens perceptions of family functioning as they relate to these traits is warranted to adequately address the exploratory questions. Practical Implications and Contribution to the Literature This study contributes to the literature by providing a more comprehensive description and analysis of family functioning pertaining to children with a DBD. The research has documented the existence of the coercive family process and its contribution to the development of antisocial behavior; however, it fails to examine the differences in family members perceptions of family cohesion and adaptability across developmental pathways. More specifically, the research has consistently proven that the manifestation of behavior problems is viewed as being directly related to dysfunctional interaction patters within the family system (Patterson et. al, 1984, 1995; Rutter, 1994; Dadds, 1997, MaddenDerdich, et. al, 2002). The persistence of such antisocial behaviors intensifies in combination with poor communication skills, ineffective conflict resolution skills, and faulty structural organization of the family system (Madden-Derdich et. al, 2002). This latter issue pertains to the lack of clarity in family roles, misinterpretation of boundaries where parents are not in a position of authority, and inability of family to accommodate to developmental and situational challenges. It is therefore
78 imperative to focus on perceptions since they offer an insiders view of what the individual is experiencing as a member of their family and sheds light on the possible breakdowns within family functioning for individual family units. Furthermore, the information gleaned from the identification of disagreements on levels of cohesion and adaptability may serve to assist in the problem identification, goal setting, and therapeutic strategies in treatment (MaddenDerdich, Leonard, and Gunnell, 2002). In summary, the inclusion of preadolescents in this study enhances the existing information about childrens perceptions of their families and encourages focused family-based intervention. It is important to note, however, that additional research is warranted to validate the childrens version of the FACES to be able to ascertain the practical significance of these findings. Limitations and Directions for Future Research Although this research study provides valuable information on perceptions of family functioning between children with a disruptive behavior disorder and their primary caregiver, there are several limitations that must be noted. First, the results can be generalized only to families with a child meeting criteria for a DBD on a teacher rating scale between the ages of 7 and 12 receiving services in an EH or SED classroom within Hillsborough or Pinellas Counties or counties that are similar. Second, there were unanticipated problems with measurement including deficient psychometric properties on the FACES and DBD-TF. In terms of the PSD, there is no reported validity data and the reliability estimates obtained in
79 the current study were extremely poor. Given that the reliability coefficient provides the percent of variance in an observed variable that is accounted for by true scores on the underlying construct, one possible interpretation that can be made with the current data is that the coefficient is projecting measurement noise. In other words, the current study may not be obtaining a true measure of CU traits. This certainly may have contributed to the non-significant relationship found between these traits and perceptions of family functioning. In examining the results of the reliability estimates gleaned from this analysis, it was evident that the coefficients of the cohesion subscale on caregiver version of the FACES was higher as compared to the Childrens Version while the trend for the adaptability subscale was reversed. The overall coefficients for the two versions were commensurate. The principal investigator noted that, in general, caregivers had a tendency to respond in a more favorable manner, thus portraying their family in a positive light. An overall comparison also was conducted between the internal consistency reliability estimates found in the current study and those reported in the literature for both the FACES-III and PSD, revealing a notable difference. More specifically, Olsen, Portner, and Bell (1982) as well as Frick, Bowen and Barry (2000) reported reliability estimates that were an average of 19 points higher (range of 5 to 41) for the two measures. Several hypotheses can be generated to explain the variability between the analyses in the current study and those found in the literature, including the differences between the participants in the current study as compared to the normative sample in that the latter group
80 was obtained from a clinical setting. In addition to the dissimilarity in sample composition, the present study consisted of a low sample size. A combination of these factors may have contributed to the differences in the reliability estimates found between the literature and the current analysis. Restriction of range on the CU traits variable, as well as measurement error also play a role, making it difficult to confidently measure the strength of the relationship between CU traits and perceptions of family functioning. Future research should further explore the relationship between CU traits and perceptions of family functioning among a more appropriate population. Since it can be hypothesized that these children are not typically served in a classroom setting in a regular elementary school, the search should be expanded to a clinical or residential type facility. A third limitation is the self-report nature of the study, which elicits concerns regarding the issue of social desirability effects for parent ratings. The accuracy of the reports are unverifiable, making it possible for the primary caregiver to portray their child or family in the way they want them to be viewed. Relatedly, the teacher ratings may be somewhat inflated and may be a reflection of their level of tolerance of behavior problems within the classroom or engaging in a self-fulfilling prophecy (i.e., the children will rate high on the rating scale because they are being served in a behavior disorders classroom). This further taps into the concept of perception as the current study examines the view of the classroom teacher only in relation to severity of behavior problems and presence of CU traits. This should be taken into consideration when interpreting the
81 results, with the understanding that parental perceptions of the childs behavior may differ due to varying expectations within the school and home settings. While the current study has expanded upon the literature by obtaining developmental perspectives in an attempt to illuminate the impact of family functioning in family units that have a child exhibiting behavioral problems, several avenues need further exploration. First and foremost, the childrens version of the FACES needs to be validated. As mentioned previously, there is no anchor with which to reliably discuss the implications of the data gathered since current interpretations are based on adolescent norms. Second, reported differences may indicate a need for further assessment regarding the relationship between the caregiver and the child with a DBD. One suggested method for addressing this is to obtain the perceptions of other family members such as siblings and other caregivers. Varying perceptions (e.g., child meeting the criteria for a DBD views family as less cohesive and adaptive while sibling views the family as more cohesive and adaptive) may support the aforementioned theory whereas similar reports (e.g., both siblings view family as less cohesive and adaptive) of perception could guide future research in the area of resilience to explain why the sibling has avoided developing a behavior disorder despite a problematic family environment. Based on the strong correlation in the literature linking family interaction and home environment to the development of behavior problems, it might be expected that all children within the family unit experience ODD or CD. If the opposite is found to be the case, then additional exploration is warranted to identify the possible protective factors in the siblings life.
82 In conclusion, this study has revealed that preadolescent children with a disruptive behavior disorder consistently produced lower scores than their caregivers on the Cohesion and Adaptability scales of the FACES. The meaning of this will have to await validation of the childrens version of the measure. Preliminarily, the findings support previous research conducted with adolescents, therefore, suggesting that DBD plays a role in the negative perceptions that children report. As discussed previously, additional research is warranted to address this hypothesis, which should include perceptions from multiple family members and inclusion of typically developing preadolescents. Overall, the current research study has provided the next stepping-stone in understanding ODD and CD as it relates to the family in that differences between perceptions have been revealed; however, the extent of these differences need further exploration to address the usefulness of the childrens version of the FACES and practical implications of child-caregiver differences.
83 References Abidin, R. R., Jenkins, C. L., McGaughey, M. C. (1992). The relationship of early family variables to childrens subsequent behavioral adjustment. Journal of Clinical Child Psychology, 21 (1), 60-69. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders Text Revision(4 th ed.). Washington, DC: American Psychiatric Association. Baden, A. D., & Howe, G. W. (1992). Mothers Attributions and expectancies regarding their conduct disordered children. Journal of Abnormal Child Psychology, 20 (5), 467-485. Barkley, R.A. (1998). Attention-Deficit Hyperactivity Disorder: A Handbook for Diagnosis and Treatment (2 nd ed). New York, NY: The Guilford Press. Behar, D., & Mark, A. (1984). Aggressive conduct disorder: The influence of social class, sex and age on the clinical picture. Journal of Child Psychology & Psychiatry & Allied Disciplines, 25 (1), 119-124. Berk, L. E. (2003). Child Development (6 th Ed.). Boston: Allyn & Bacon. Combrinck-Graham, L. (1990). Developments in family systems theory and research. Journal of the American Academy of Child and Adolescent Psychiatry, 29 (4), 501-512.
84 Dadds, M. R. (1987) Families and the origins of child behavior problems. Family Process, 26 (3), 341-357. Dishion, T. J., French, D. C., Patternson G. R. (1995). The development and ecology of antisocial behavior. In Cicchetti, D. & Cohen, D. J. (Eds.). Developmental psycholpathology, Vol. 2: Risk disorder, and adaptation (pp. 421-471). New York, NY: John Wiley & Sons, Inc. Fisher, S. W. (1996). The family and the individual: Reciprocal influences. In Vanzetti, N., Duck, S. (Eds.), A lifetime of relationships (pp. 311-335). Pacific Grove, CA: Brooks/Cole Publishing Co. Frick, P. J. (2000). A developmental psychopathology approach to understanding and preventing youth violence. Journal of the Louisiana State Medical Society, 152, 497-502. Frick, P. J. (1993). Childhood conduct problems in a family context. School Psychology Review, 22 (3), 376-385. Frick, P.J., Boden, S.D., and Barry, C.T. (2000). Psychopathic traits and conduct problems in community and clinic-referred samples of children: Further development of the psychopathy screening device. Psychological Assessment, 12 (4), 382-393. Frick, P. J., Christian, R. E., Wooton, J. M. (1999). Age trends in association between parenting practices and conduct problems. Behavior Modification, 23 (1), 106-128.
85 Frick, P. J., & Ellis, M. (1999). Callous-Unemotional traits and subtypes of conduct disorder. Clinical Child and Family Psychology Review, 2 (3), 149-168. Frick, P. J., Lahey, B. B., Hrtdagen, S., & Hynd, G. W. (1989). Conduct problems in boys: Relations to maternal personality, marital satisfaction, and socioeconomic status. Journal of Clinical Child Psychology, 18 (2), 114-120. Gardner, F. E. M. (1992). Parent-child interaction and conduct disorder. Educational Psychology Review, 4 (2), 135-163. Gelfand, D.M., Jenson, W. R., & Drew, C. J. (1982). Understanding child behavior disorders: An introduction to child psychopathology. (2 nd ed, pp. 236-251). Ghuman, H. S. (1998). Conduct and oppositional defiant disorders. In Ghuman, H. S., & Sarles, R. M. Handbook of child and adolescent outpatient, day treatment and community psychiatry. ( pp. 175-185). Philadelphia, PA: Brunner/Mazel, Inc. Haddad, J. D., Barocas, R., & Hollenbeck, A. R. (1991). Family organization and parent attitudes of children with conduct disorder. Journal of Clinical Child Psychology, 20 (2), 152-161. Hatcher, L. & Stepanski E., J. (1999). A Step-by-Step Approach to Using the SAS System for Univariate and Multivariate Statistics. Cary, NC: SAS Institute Inc.
86 Johnson, B. (2001). Toward a new classification of nonexperimental quantitative research. Educational Researcher, 30 (2), 3-13. Kazdin, A. E. (1993). Treatment of conduct disorder: Progress and directions in psychotherapy research. Development and Psychopathology, 5 277-310. Luthar, S., Burack J., Cicchetti, D. & Weisz, J. (1997). Developmental psychopathology: Perspectives on adjustment, risk, and disorder. New York, NY: John iley & Sons, Inc. Lyons-Ruth, K. (1996). Attachment relationships among children with aggressive behavior problems: The role of disorganized early attachment patterns. Journal of Consulting and Clinical Psychology, 64 (1), 64-73. Madden-Derdich D. A., Leonard, S. A., & Gunnell G. A. (2002). Parents and childrens perceptions of family processes in inner-city families with delinquent youths: A qualitative investigation. Journal of Marital and Family Therapy, 28 (3), 355-369. Mash, E.J. & Barkley, R. A. (1996). Child Psychopathology. New York, NY: The Guilford Press. Mikesell, R. H., Lusterman, D., & McDaniel, S. H. (1995). Integrating family therapy: Handbook fo family psychology and systems theory. Washington, DC: American Psychological Association. Noller, P. & Callan, V. J. (1986). Adolescent and parent perceptions of family cohesion and adaptability. Journal of Adolescence, 9, 97-106.
87 Noller, P., Seth-Smith, M., Bouma, R., & Schweitzer, R. (1992). Parent and adolescent perceptions of family functioning: A comparison of clinic and non-clinic families. Journal of Adolescence, 15 (2), 101-114. Olson, D. H. (1982). Family Inventories Manual. Minneapolis, MN: Life Innovations. Patterson, G. R., Dishion, T. J., & Bank, L. (1984). Family interaction: A process model of deviancy training. Aggressive Behavior, 10 (3), 253-267. Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). Antisocial boys. Eugene, OR: Castalia. Pelham, W.E., Gnagy, E. M., Greenslade, K.E., Milich, R. (1992). Teacher rating of DSM-III-R symptoms for the disruptive behavior disorders. Journal of American Academic Child and Adolescent Psychiatry, 31 (2), 210218. Pillay, A. L. (1998). Perceptions of family functioning in conduct disordered adolescents. South African Journal of Psychology, 28 (4), 191-195. Robinson, E. A. (1985). Coercion theory revisited: Toward a new theoretical perspective on the etiology of conduct disorders. Clinical Psychology Review, 5, 597-625. Rutter, M. (1994). Family discord and conduct disorder: Cause, Consequence, or correlate? Journal of Family Psychology, 8 (2), 170-186. Rutter, M. (1985). Family and school influences on behavioural development. Journal of Child Psychology & Psychiatry & Allied Disciplines, 26 (3), 349368.
88 Short, R. J. & Shapiro, S. K. (1993). Conduct disorders: A framework for understanding and intervention in schools and communities. School Psychology Review, 22 (3), 362-375. Slee, P. T. (1996). Family climate and behavior in families with conduct disordered children. Child Psychiatry and Human Development, 26 (4), 255-265. Steiner, H. & Wilson, J. (1999). Conduct Disorder. In Hendren, R. et al., Disruptive behavior disorders in children and adolescents. Washington, DC: American Psychiatric Press. Steiner, H. (1997). Practice parameters for the assessment and treatment of children and adolescents with conduct disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 36 (10S), Supplement: 122S139S. Szapocznik, J., Rio, A., Murray, E., Cohen, R., Scopetta, M., Rivas-Vasquez, A., Hervis, O., Posada, V., & Kurtines, W. (1989). Structural family versus psychodynamic child therapy for problematic Hispanic Boys. Journal of Consulting and Clinical Psychology, 57, 571-578 Thomas, A. & Chess, S. (1984). Genesis and evolution of behavioral disorders: From infancy to early adult life. American Journal of Psychiatry, 141, 1-9. Wooton, J. M., Frick, P. J., Shelton, K. K., Silverthorn, P. (1997). Ineffective parenting and childhood conduct problems: The moderating role of callous-emotional traits. Journal of Consulting and Clinical Psychology, 65 (2), 292-300.
90 APPENDIX A Family Information Child Information: Childs Name: ________________________ ID#: _____________ Childs Age: ____ Childs Grade Level: _____ Childs Gender: ___ Ethnicity: _______ African-American _______ Hispanic _______ Asian _______ Native American _______ Caucasian _______ Other: _________ Primary Caregiver Information Relationship to child: ______________________________________ Age: ____________ Ethnicity: _____________________________ Level of Education: __________________ Members living in household and their relationship to child:
91 APPENDIX B Parent/Teacher Disruptive Behavior Disorders Rating Scale Childs Name: ______________________ Form Completed By: ___________ Grade: ____ Date of Birth: ________ Sex: ___ Date Completed: ________ Check the column that best describes your/this child. Please write DK next to any items for which you do not know the answer. Not at All Just a Little Pretty Much Very Much 1. Often interrupts or intrudes on others (e.g., butts into conversations or games) 2. Has run away from home overnight at least twice while living in parental or parental surrogate home (or once without returning for a lengthy period) 3. Often argues with adults 4. Often lies to obtain goods or favors or to avoid obligations (i.e., cons others) 5. Often initiates physical fights with other members of his or her household 6. Has been physically cruel to people 7. Often talks excessively 8. Has stolen items of nontrivial value without confronting a victim (e.g., shoplifting, but without breaking and entering, forgery) 9. Is often easily distracted by extraneous stimuli 10. Often engages in physically dangerous activities without considering possible consequences (not for the purpose of thrillseeking), e.g., runs into street without looking 11. Often truant from school, beginning before age 13 years 12. Often fidgets with hands or feet or squirms in seat 13. Is often spiteful or vindictive 14. Often swears or uses obscene language 15. Often blames others for his or her mistakes or misbehavior. 16. Has deliberately destroyed others property (other than by fire setting)
92 APPENDIX B (Continued) Not at All Just a Little Pretty Much Very Much 17. Often blurts out answers before questions have been completed. 18. Often initiates physical fights with others who do not live in his or her household (e.g., peers at school or in the neighborhood) 19. Often shifts from one uncompleted activity to another 20. Often has difficulty playing or engaging in leisure activities quietly 21. Often fails to give close attention to details or makes careless mistakes in schoolwork, work, or other activities 22. Is often angry and resentful 23. Often leaves seat in classroom or in other situations in which remaining seated is expected 24. Is often touchy or easily annoyed by others 25. Often does not follow through on instructions and fails to finish school work, chores, or duties in the workplace (not due to oppositional behavior or failure to understand instruction) 26. Often loses temper 27. Often has difficulty sustaining attention in tasks or play activities 28. Often has difficulty awaiting turn 29. Has forced someone into sexual activity 30. Often bullies, threatens, or intimidates others 31. Is often on the go or often acts as if driven by a motor 32. Often loses things necessary for tasks or activities (e.g., toys, school assignments, pencils, books, or tools) 33. Often runs about or climbs excessively in situations in which it is inappropriate 34. Has been physically cruel to animals 35. Often avoids, dislikes, or is reluctant to engage in tasks that require sustained mental effort (such as schoolwork or homework)
93 APPENDIX B (Continued) Not at All Just a Little Pretty Much Very Much 36. Often has difficulty organizing tasks and activities 37. Has broken into someone elses house, building, or car 38. Is often forgetful in daily activities 39. Has used a weapon that can cause serious physical harm to others (e.g., bat, brick, broken bottle, knife, gun)
95 APPENDIX C FACES II: Family Version David H. Olson, Joyce Portner & Richard Bell 1 Almost Never 2 Once in a While 3 Sometimes 4 Frequently 5 Almost Always 1. Family members are supportive of each other during difficult times. 2. In our family, it is easy for everyone to express his/her opinion. 3. It is easier to discuss problems with people outside the family than with other family members. 4. Each family member has input regarding major family decisions. 5. Our family gathers together in the same room. 6. Children have a say in their discipline. 7. Our family does things together. 8. Family members discuss problems and feel good about the solutions. 9. In our family, everyone goes his/her own way. 10. We shift household responsibilities from person to person. 11. Family members know each others close friends. 12. It is hard to know what the rules are in our family. 13. Family members consult other family members on personal decisions. 14. Family members say what they want.
96 APPENDIX C (Continued) 1 Almost Never 2 Once in a While 3 Sometimes 4 Frequently 5 Almost Always 15. We have difficulty of thinking of things to do as a family. 16. In solving problems, the childrens suggestions are followed. 17. Family members feel very close to each other. 18. Discipline is fair in our family. 19. Family members feel closer to people outside the family than to other family members. 20. Our family tries new ways of dealing with problems. 21. Family members go along with what the family decides to do. 22. In our family, everyone shares responsibilities. 23. Family members like to spend their free time with each other. 24. It is difficult to get a rule change in our family. 25. Family members avoid each other at home. 26. When problems arise, we compromise. 27. We approve of each others friends. 28. Family members are afraid to say what is on their minds. 29. Family members pair up rather than do things as a total family. 30. Family members share interests and hobbies with each other.
97 APPENDIX D FA C ES II C hildrens V ersion (Modified from FACES-II) Describe your family: 1. In my family, we help each other. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 2. In my family, it is easy to tell each other what we think. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
98 APPENDIX D (Continued) 3. When I feel bad, it is easier to talk to people outside my family. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 4. Family decisions are made together. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 5. We often all hang out with each other in the same room. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
99 APPENDIX D (Continued) 6. Kids in my family help choose their own punishment. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 7. My family does things together. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 8. When there is trouble in my family, we talk about it together and come to a solution we are happy about. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
100 APPENDIX D (Continued) 9. In my family, everyone does their own thing. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 10. We take turns doing chores. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 11. We all know each others friends. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
101 APPENDIX D (Continued) 12. It is hard to know what the rules are in our family. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 13. In my family, we talk together before deciding to do something. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 14. Everyone in my family says what they want to say. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
102 APPENDIX D (Continued) 15. It is hard for my family to think of things to do together. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 16. When there is trouble in my family, parents listen to what the kids have to say. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 17. People in my family feel very loving towards each other. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
103 APPENDIX D (Continued) 18. The punishments are fair in my family. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 19. In my family, we feel closer to people outside the family than to each other. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 20. In my family, we try new ways of helping each other. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
104 APPENDIX D (Continued) 21. In my family, we all do what the family decides to do. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 22. Everyone in my family helps with the chores. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 23. In my family, we like to spend our free time with each other. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
105 APPENDIX D (Continued) 24. The rules in my family never change. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 25. In my family, we stay away from each other at home. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 26. When people in my family disagree, we talk about it and decide on a solution that everyone agrees with. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
106 APPENDIX D (Continued) 27. We like each others friends in my family. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 28. In my family, people are afraid to say what they want. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot 29. Family members go off and do things separately instead of doing things together as a whole family. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
107 APPENDIX D (Continued) 30. In my family, we talk to each other about our hobbies and activities. No Never No Sometimes Yes, Sometimes No Yes Yes A Lot
108 APPENDIX E Psychopathy Screening Device (Parent Version) Name of Child : ______________________ Date of Birth : ___/___/____ Completed By : Mother Father Other: __________________ Date Completed : ____/_____/_____ Not at all True Sometimes True Definitely True 1. Blames other for his/her mistakes 0 1 2 2. Engages in illegal activities 0 1 2 3. Is concerned about how well he/she does at school/work. 0 1 2 4. Acts without thinking of the consequences 0 1 2 5. His/her emotions seem shallow and not genuine. 0 1 2 6. Lies easily and skillfully. 0 1 2 7. Is good at keeping promises. 0 1 2 8. Brags excessively about his/her abilities, accomplishments, or possessions. 0 1 2 9. Gets bored easily. 0 1 2 Instructions : Please complete the background information above. Then read each statement and decide how well it describes your child. Mark your answer by circling the appropriate number (0-2) for each statement. Do not leave any statement unrated.
109 APPENDIX E (Continued) 10. Uses or cons other people to get what he/she wants. 0 1 2 Not at all True Sometimes True Definitely True 11. Teases or makes fun of other people. 0 1 2 12. Feels bad or guilty when he/she does something wrong. 0 1 2 13. Engages in risky or dangerous activities. 0 1 2 14. Can be charming at times, but in ways that seem insincere or superficial. 0 1 2 15. Becomes angry when corrected or punished. 0 1 2 16. Seems to think that he or she is better or more important than other people. 0 1 2 17. Does not plan ahead, or leaves things to the last minute. 0 1 2 18. Is concerned about the feelings of others. 0 1 2 19. Does not show feelings or emotions. 0 1 2 20. Keeps the same friends. 0 1 2
110 APPENDIX F Parental Information Letter My name is Melissa Todd, and I am a graduate student at the University of South Florida in the Department of Psychological and Social Foundations. I am doing a research study at your childs school and would like to ask for your help. The goal of my study is to learn more about how caregivers and children differ in their views of their family. Child behavior problems can be very challenging for a family and additional research on the impact of the family is crucial in the development of interventions. In order to reach my goal, I will need to gather information from students and their primary caregiver. Your child has been selected to participate in this study as a result of his/her enrollment within your childs school. Your childs participation in this study will last for approximately 20 minutes and will take place in your childs classroom. The process will begin by explaining to your child that his/her participation is voluntary and that his/her answers will not be graded or shared with others (i.e., teacher, peers, etc.). Next, a graduate student will administer one questionnaire, reading each item aloud. At the end of the questionnaire administration, your child will receive a treat (e.g., candy, pencils) and be thanked for his or her participation. Your childs teacher also will be asked to complete two questionnaires that examines the presence of behavior problems as well as personality traits. In addition to obtaining the childrens view of their family, the perspective of the primary caregiver also is needed. A phone interview will be conducted with you that will last approximately 5-10 minutes. The interview will begin by gathering demographic information such as age, gender, ethnicity, and number of people in household. A questionnaire will then be administered to obtain information regarding your perceptions of your family Upon completion of the interview, you will receive a $10.00 money order for your participation and time. The education and services your children are receiving will not be affected as a result of their participation. Further, there are no foreseeable risks involved for participating. Aside from the treat your children will receive for participation, they will not directly benefit from participation in this study. All information that is gathered from you and your child will be confidential. Your family will be assigned a number, which will be placed on each questionnaire. The information will be used for research purposes only and will not be shared with your childrens teacher or school. Authorized research personnel, employees of the Department of Health and Human Services and the USF Institutional Review Board may inspect your records from this research project. The results of the study may be published in grouped form. In other words, the published results will not include your child's name or any other information that will identify you or your child. Please understand that participation in this study is completely voluntary and that you and/or your child may decide to not participate at any time. If you or your child wish to
111 APPENDIX F (Continued) not participate or change your mind at any time, your childs education will in no way be effected. If you have any questions regarding my study, please call Melissa Todd at (813) 9914034 or Linda Raffale Mendez, Ph.D. at (813) 974-1255. If you have any questions about your rights and your childs rights as a person taking part in a study, you may contact the Division of Research Compliance at (813) 974-5638.
112 APPENDIX G. Your Consent By signing this form I agree that: I have fully read or have had read and explained to me in my native language this informed consent form describing a research project. I have had the opportunity to question one of the persons in charge of this research and have received satisfactory. I understand that I am being asked to allow my child to participate in research. I understand the risks and benefits and I freely give my consent to participate in the research project outlined in this form, under the conditions indicated in it. I have been given a signed copy of this informed consent form, which is mine to keep. ________________________________________ Signature of Parent or Guardian _________________________________________ Printed Name of Parent or Guardian Investigator Statement I certify that participants have been provided with an informed consent form that has been approved by the University of South Florida's Institutional Review Board. That contains the nature, demands, risks and benefits involved in participating in this study. I further certify that a phone number has been provided in the event of additional questions. ___________________________________________ Signature of Investigator Printed Name of Investigator Date Institutional Approval of Study and Informed Consent This research project/study and informed consent form were reviewed and approved by the University of South Florida Institutional Review Board for the protection of human subjects. This approval is valid until the date provided below. The board may be contacted at (813) 974-5638.
112 APPENDIX H Child Assent Form I want to be a part of the study taking place at my school. An adult explained the study to me. I understand that it is about how I feel about my family. I was told that I will be asked to fill out one questionnaire and that an adult will read each question out loud to the class. I understand that I can change my mind at any time. I know that I will receive a treat once all the questions have been answered. ______________________________ Your Name ______________________________ Date ______________________________ Witness (Print name) ______________________________ Witness (Sign name)
113 APPENDIX I Instructions for Group Administration (Children) Hi! Thank you for agreeing to help out with this project. Today we are going to ask you several questions about your feelings towards other family members and the rules in your household. Listen carefully to each statement and answer as honestly as you can we will not share any of your answers with anyone, so your parents and teachers will have no idea what you have written on your papers. For each statement I want you to think about your family and how you would best describe it. The answers to chose from are: (1) almost never, (2) once in a while, (3) sometimes, (4) very often, and (5) almost always. Lets practice. Read the following sample item to children: I like to watch television on Saturday mornings. Then say: Think about whether or not this statement is true for you. If you like to watch TV every Saturday morning or almost every Saturday morning, then you would choose "almost always." If you like to watch TV on Saturday mornings many Saturdays per month (3) then you would respond very often. If you sometimes like to watch TV on Saturday mornings (e.g., twice a month), then you would pick "sometimes." If you only like to watch TV on Saturday mornings every few weeks, say only 1 Saturday a month, then you would choose once in a while. Finally, almost never would mean that you hardly ever watch TV on Saturday mornings. Does anyone have any questions before we start? Once you have finished you will receive a treat for participating.
114 APPENDIX J Debriefing Protocol Children Thank you for helping out with this study. The purpose of asking you about your family was to better understand families and how different family members think about things such as the rules and feelings towards each other. All families are different in their own way and we want to learn about these differences. Does anyone have any questions? Primary Caregiver Thank you for volunteering your time for this interview. The purpose of the study is to learn more about families in relation to how they perceive the roles of each family member as well as how close family members feel toward each other. The information collected in this study will also help us to determine if children with behavior problems view their family differently. We can use this information to develop family-based interventions.
115 APPENDIX K May 23, 2003 Dear Thank you for participating in my research study regarding the perceptions of family functioning among children and caregivers. Your input was valuable to me and I appreciate you sharing your thoughts and opinions. As mentioned during our phone conversation, I am providing you with a $10.00 money order as a way to express my gratitude. If you have any additional questions, please do not hesitate to contact me at 813-390-7698. Many thanks, Melissa Todd, M.A.