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 001787973
007 cr mnu|||uuuuu
008 070515s2005 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001435
Curley, Jessica K.
Mothers' versus Fathers' ratings of child behavior problems
h [electronic resource] /
by Jessica K. Curley.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: The goal of this study was to examine how mothers and fathers view children's internalizing and externalizing behavior problems. More specifically, the relationship between certain factors, such as parental psychological symptoms, levels of interparental conflict, characteristics of the behaviors, and discrepancies in mothers' and fathers' ratings of behavior problems were studied in more depth. Using a between subjects, experimental design, mothers and fathers were randomly assigned to view a videotape and rate the behavior of a male or female child acting in either an internalizing, externalizing, or non-clinical manner. Results showed that there were no differences between mothers' and fathers' ratings of the videos and the parents' own psychological symptoms and interparental conflict were not associated with higher ratings of the child in the videos. However, main effects were found for the type of video that the participant watched and the gender of the child in the video. In addition, interactions between the type of video and the gender of the child in the video were found for ratings on the Anxious-Depressed, Withdrawn-Depressed, and Aggressive Behavior subscales of the Child Behavior Checklist (CBCL). In order to explain the present findings, level of contact with children, child socialization, and gender roles were explored in further depth.
Thesis (M.A.)--University of South Florida, 2005.
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 85 pages.
Adviser: Vicky Phares, Ph.D.
t USF Electronic Theses and Dissertations.
MothersÂ’ Versus FathersÂ’ Ratings of Child Behavior Problems by Jessica K. Curley A thesis submitted in partial fulfillment of the requirement s for the degree of Master of Arts Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Vicky Phares, Ph.D. Tammy Allen, Ph.D. Ellis Gesten, Ph.D. Date of Approval: December 14, 2005 Keywords: parents, children, psychologic al symptoms, interparental conflict, gender Copyright 2006, Jessica K. Curley
Dedication First, I would like to dedicate this thesis to my family who without them I would not have had the courage and strength to get to where I am today. Also, I want to thank them for their unconditional suppor t throughout this project. In addition, I would like to dedicate this thesis to my boyfriend, Clay, for his encouragement through the stressful times and always lend ing a sympathetic ear. Lastly, I would like to dedicate this thesis to my major professor, Vicky Phares, whose knowledge, inspiration, and constant s upport have greatly impacted my graduate career thus far.
i Table of Contents List of Tables ........................................................................................................ iii List of Figures .......................................................................................................v Abstract ................................................................................................................vi Introducti on...........................................................................................................1 Amount of Contac t.....................................................................................2 Parental Psycholog ical Symp toms.............................................................4 Interparental Conflict and Family Distress.................................................9 Types of Child Behavior...........................................................................11 Gender Diffe rences ..................................................................................12 The Presen t Study...................................................................................14 Method ...............................................................................................................17 Partici pants..............................................................................................17 Measures.................................................................................................20 Interview/Videot ape Stimul us........................................................20 Child B ehavior ...............................................................................20 Parental Psycholog ical Symp toms................................................21 Interparental Confli ct.....................................................................22 Demographics and Time S pent.....................................................23 Procedur es..............................................................................................23 Development of Vi deos and Pilot Study........................................23 Present Study................................................................................26 Results ...............................................................................................................30 Randomization and Relia bility Ch ecks.....................................................30 Differences Between Mothers and Fathers on Vid eo Ratings..................31 ParentsÂ’ Psychological Sympto ms, Interparental Conflict, and Own Child Ratings ........................................................................44 Discussio n..........................................................................................................49 Lack of Mother-Father Diffe rences in Ra tings..........................................49 Lack of Influence of Psychological Symptoms and Interparental Conflic t.....................................................................................................52 Child Gender Differenc es.........................................................................55 Limitations and Futu re Resear ch.............................................................61
ii Conclusi ons.............................................................................................63 List of Refe rences...............................................................................................65 Appendice s.........................................................................................................71 Appendix A: Sample questions from Structured Clinical Interview For Children and Adol escents (SCI CA)..............................72 Appendix B: Selected Child B ehavior Checkl ist Items.............................73 Appendix C: OÂ’LearyPorter Sc ale...........................................................75 Appendix D: Parental Demographic Form...............................................76 Appendix E: Questionnai re for Ra ters.....................................................78 Appendix F: Additi onal Rating Form........................................................81 Appendix G: Info rmed Cons ent................................................................82 Appendix H: Debr iefing Fo rm...................................................................85
iii List of Tables Table 1. MothersÂ’ and fathersÂ’ demographic va riables................................19 Table 2. Analysis of variance a nd chi-square test s to determine whether random assignment ac ross videos was effective............31 Table 3. MothersÂ’ and fathersÂ’ descr iptive statistics of their mean ratings averaged acro ss the six vi deos.........................................32 Table 4. Multivariate and univariate F values for Parent Gender by Type of Video by Child Gender of Video interactions for parentsÂ’ ratings of the video on the four subscales of the CBCL............................................................................................34 Table 5. Tukey post hoc tests for the Type of Video by Child Gender interaction across the Anxi ous-Depressed, WithdrawnDepressed, and Aggressive B ehavior subscale scores................37 Table 6. Multivariate and univariate F values for Parent Gender by Type of Video by Child Gender of Video interactions for the Internalizing Mean, Externaliz ing Mean, and Total Mean of the parentsÂ’ ratings ........................................................................41 Table 7. MothersÂ’ descriptive stat istics for Brief Symptom Inventory (BSI), OÂ’Leary Porter Scale (O PS), and their ratings of their own child on the Child Behav ior Checklist (CBCL)........................45 Table 8. FathersÂ’ descriptive stat istics for Brief Symptom Inventory (BSI), OÂ’Leary Porter Scale (O PS), and their ratings of their own child on the Child Behav ior Checklsist (CBCL)......................45 Table 9. Multiple regression analyses of mothersÂ’ and fathersÂ’ ratings on the Brief Symptom Inventor y (BSI) and OÂ’Leary Porter Scale (OPS) predicting ratings of the videos across four behavior subscale s......................................................................................46 Table 10. Multiple regression analyses of mothersÂ’ and fathersÂ’ ratings on the Brief Symptom Inventor y (BSI) and OÂ’Leary Porter Scale (OPS) predicting ratings of t he videos across Internalizing Mean, Externalizing Mean, and Tota l Mean ..................................46
iv Table 11. Pearson correlations of parent sÂ’ ratings of the child in the video on items from four subsca les of the CBCL (Internalizing Mean, Externalizing Mean, and Total Mean) and the parentsÂ’ ratings of their child closest in age to 6 on the CBCL (Internalizing T-score, Exte rnalizing T-score, and Total T-score).........................................................................................48
v List of Figures Figure 1. Interaction between type of video and child gender for participantsÂ’ ratings on the Anxious-Depressed subscale of the CBCL.................................................................................. 38 Figure 2. Interaction between type of video and child gender for participantsÂ’ ratings on the Anxious-Depressed subscale of the CBCL.................................................................................. 39 Figure 3. Interaction between type of video and child gender for participantsÂ’ ratings on the Aggressive Behavior subscale of the CBCL.................................................................................. 40 Figure 4. Interaction between type of video and child gender for participantsÂ’ ratings of the Exte rnalizing Mean (items from both externalizing scale s of the CB CL).........................................43
vi MothersÂ’ Versus FathersÂ’ Ratings of Child Behavior Problems Jessica K. Curley ABSTRACT The goal of this study was to exam ine how mothers and fathers view childrenÂ’s internalizing and externalizing behavior problems. More specifically, the relationship between certain factor s, such as parental psychological symptoms, levels of inter parental conflict, characteri stics of the behaviors, and discrepancies in mothersÂ’ and fathersÂ’ ra tings of behavior problems were studied in more depth. Using a between subj ects, experimental design, mothers and fathers were randomly assigned to view a videotape and rate the behavior of a male or female child acting in either an in ternalizing, externalizing, or non-clinical manner. Results showed that there were no differences between mothersÂ’ and fathersÂ’ ratings of the videos and the par entsÂ’ own psychological symptoms and interparental conflict were not associated with higher ratings of the child in the videos. However, main effects were found for the type of video that the participant watched and the gender of the child in t he video. In addition, interactions between the type of video and the gender of the child in the video were found for ratings on the Anxious-D epressed, Withdrawn-Depressed, and Aggressive Behavior subscales of the Child Behavior Checklist (CBCL). In order to explain the present findings, level of c ontact with children, ch ild socialization, and gender roles were explored in further depth.
1 Introduction Clinicians rely heavily upon the ratings of parents and teachers when assessing children for behavior problems. However, one of the difficulties in gathering information from informants on ch ildren is that the ratings of behavior problems often differ markedly. Ac henbach, McConaughy, and Howell (1987) proposed that these differences may be due to situational specificity because childrenÂ’s behavior and emotional probl ems span a wide range of situations, such as at school, home, clinic, and neigh borhood. Therefore, raters who see these children in different environments may differ in their ratings of internalizing, externalizing, and overall beh avior problems. Children have been known to act graciously and obedient at school, but then act out at home perhaps due to a lack of structure set fort h by parents. The opposite can be true as well. The correlation between different informant s on childrenÂ’s behavior in similar situations is .60. However, info rmantsÂ’ agreement on behavior in different situations, such as home and school is mu ch lower, averaging about .28. The correlation for informants in similar situat ions is relatively high but it still only accounts for less than 40% of the variance in explaining child renÂ’s behavior. Thus, it is important to gather informati on from more than one informant about a childÂ’s behavior (Achenbach et al., 1987). Leaving teachers, mental health work ers, and peers aside, mothers and fathers often differ in their ratings of their childrenÂ’s behaviors. Given that the
2 large majority of interactions between parents and children occur in the home, situational specificity shoul d not account for these differences. Several factors have been shown to influence agreement between mothers and fathers on childrenÂ’s internalizing and externalizi ng behavior problems. These factors include, the amount of c ontact with the child, parental psychological symptoms and personality, interparental conflict and fa mily distress, and characteristics of the behavior. Each of these fact ors will be reviewed in more depth. Amount of Contact It is commonplace to include only mo thers in research on children and to place less emphasis on fathersÂ’ roles in clinical settings (Phares, 1992; Phares & Compas, 1992; Phares, Lopez, Fields, Kamboukos, & Duhig, 2005). This process may be fueled by the fact that mothers tend to have significantly more contact with their children than do fat hers. Additionally, there has been a tendency to rely too heavily upon mother sÂ’ ratings for assessment of childrenÂ’s behavior because mothers usually have t he most contact with the child. However, researchers have cautioned again st overreliance on mothersÂ’ ratings due to the motherÂ’s own adjustment pr oblems and psychologic al symptoms that may influence ratings of behavior probl ems (Chi & Hinshaw, 2002; Christensen, Phillips, Glascow, & Johnson, 1983; Kroes Veerman, & DeBruyn, 2003). Also, mothers who do spend more time with t he child tend to report more behavior problems than fathers, especially in very young children (Achenbach, 1992). Because mothers are more involved in child rearing and since they may have more awareness and insight into thei r childÂ’s behaviors, Christensen and
3 colleagues (1992) predicted that mothers would overreport child behavior problems in comparison to fathers. Their hypothesis was supported when mothers reported a signific antly higher number of negative child behaviors than did fathers. Likewise, Webster-Stratt on (1988) found that mothers see their childrenÂ’s behavior problems as occurri ng more frequently and are more likely to perceive these problems as a threat to t heir well-being. Mothers may internalize the challenge of raising a child wi th behavior problems and may become more distressed than fathers simply because mo thers spend more time with the child. Another reason for the discrepancy between mothersÂ’ and fathersÂ’ ratings could be that children obey their fathers more frequently than their mothers and children are more likely to behave appropria tely in front of their fathers even when in the presence of their mother (Campbell, 1991). Thus, the fathersÂ’ contact time could consist of fewer and less severe child behavioral problems than mothersÂ’ contact time and consequent ly, mothers and fathers may both be portraying their experience of the childÂ’s behavior accurately. As suggested, mothersÂ’ greater like lihood of reporting negative behaviors could be due to their increased exposure to their childÂ’s behavio r. Therefore, mothers may possibly be a more accurate informant of their childÂ’s behavior. Additionally, maternal psychopathology puts children at risk for developing the psychological symptoms. Likewise, a mo ther may be accurate ly reporting higher levels of behavior problems in their ch ildren as suggested by the accuracy model (Richters, 1992). Conversely, mothersÂ’ psychological symptoms may distort their perceptions of their child Â’s behaviors, which may lead to an overreporting of
4 problem behaviors. Richters (1992) ca lled this interpretation the distortion model. Najman et al. (2000) suggested t hat a combination of the two models is true such that, depressed mothers do r eport higher levels of child behavior problems but that parental p sychopathology may lead to a real increase in child behavior problems. Since mothers have mo re contact with the child, they may provide more accurate accounts of behavio r problems but their ratings may still be distorted. The connection between di stortion and parental psychological symptoms is addressed next. Parental Psychological Symptoms Phares, Compas, and Howell (1989) f ound that the correlation between parentsÂ’ ratings of behavior problems is influenced by parental psychological symptoms. Several researchers have r eplicated this finding in mothers but evidence of the same pattern of findings with fathers is mixed. For instance, Phares et al. (1989) found that both mo thersÂ’ and fathersÂ’ reports of their psychological symptoms and their reports of their childrenÂ’s behavior were significantly associated. WebsterStratton (1988) found that mothersÂ’ psychological problems, in particular depression, were better predictors of maternal reports of child behavior problem s than teachersÂ’ repor ts of the childÂ’s negative behaviors. In contrast, fathersÂ’ reports were much less influenced by personal adjustment measures. In addi tion, significant correlations between fathersÂ’ and teachersÂ’ reports were pr esent; however, the correlation between mothersÂ’ and teachersÂ’ reports was small. Yet, the correlation between fathers and teachers was stronger for externalizi ng than for internalizing disorders in
5 children when examining the CBCL scales separately. Interestingly, mothers who were observed in the home as ex hibiting critical and physically negative behaviors had higher ratings of child behav ior problems. However, there were no significant correlations between father sÂ’ reports and paternal behaviors in the home. Kurdek (2003) studied the c onnection between personality and psychopathology in relation to parentsÂ’ ra tings of child behavior problems. For fathers, high ratings of child behavior problems were associated with personality characteristics such as high levels of neuroticism, low levels of conscientiousness, and low levels of openness. According to the big five model, individuals high in neuroticism are characterized as having difficulty controlling impulses and dealing with stress. Also neuroticism has been identified as a facet of depression (Costa & McCrae, 1992). It seems plausible that fathers with these traits would find child behavior problem s difficult. Similarly, fathers low on openness and conscientiousness would view unforeseen incidents and interferences in dealing wit h children as more problematic than those high on these personality traits. MothersÂ’ reports of frequent child behavior problems were also correlated with high levels of neuroticism but not conscientiousness. Unlike fathers, a higher frequency of r eported behavior problems was associated with higher levels of openness. Higher levels of openness were characterized as being unconventional and as experiencin g positive and negative emotions more intensely. Therefore, mothers with high levels of openness may be unlike conventional mothers who see childrenÂ’ s behavior problems as normal and may
6 be more distressed due to thei r experience of negative emot ions. In short, both fathersÂ’ and mothersÂ’ personal ity traits and negative affe ct are related to their reports of their child behavior problems. Other researchers have explored the association between maternal depression and high ratings of child behavio r problems. In particular, Chi and Hinshaw (2002) investigat ed the depression distortion hypothesis (Richters, 1992) and found that mothersÂ’ depressive symptoms predicted elevated ratings of child ADHD symptoms and contributed to negative biases in reports of their own parenting behavior. This finding t hat the distortions transcended the maternal reports of their childrenÂ’s behaviors and incorporat ed views of the mothersÂ’ own parenting was unprecedented. Additionally, these biased maternal ratings were even higher than teachersÂ’ r eports on hyperactivity, inattentiveness, and disruptiveness. One important limitation to research is the elucidation of this depression-distortion hypothesis. Un less experimental m anipulations are performed, it will be difficult to assess t he accuracy of informants. Child and teacher ratings were used as criterion for maternal ratings in the Chi and Hinshaw (2002) study, but because child behaviors differ across situations, one can only infer who is most accurate in each case (Achenbach et al., 1987). Overall, if elevated emotional distre ss and depressive symptoms in maternal raters are present, then assessment in formation should be evaluated with care, and when possible multiple informants s hould be accessed for information on the childÂ’s behavior.
7 A study by Kroes, Veerman, and De Bruyn (2003) also investigated how to interpret high ratings of problematic ch ild behavior reported by mothers with high levels of psychopathology. This study again looked at RichtersÂ’ (1992) two competing interpretations, the distor tion and the accuracy models. Growing research has shown that parental psyc hopathology is related to emotional and behavioral problems in their childr en (accuracy model) and that parental psychopathology leads to distortions in parental reports (dis tortion model; Kroes et al., 2003). Multiple regression analyse s were used to show the amount of variance in reports due to corres pondence between mothers and teachers (reflecting accuracy) and the amount of variance due to maternal psychopathology (reflecting distortion) MothersÂ’ symptomatology had a significantly greater distortion effect on t he reports of internalizing child behavior problems than for externalizing behavior problems. This distortion was also related to the types of maternal psychopat hology. For instance, maternal hostility produced distortions in reporting of exter nalizing behavior problems, but maternal depression did not. The distortion model does appear to be re lated to maternal ratings and the authors suggested some alternative explanat ions for this association. These were the projection hypothesis (Moretti, Fine, Haley, & Marriage, 1985) and the social attribution theory (Dodge, 1986). The projection hypothes is states that mothers project their own sym ptoms onto their children. This assumption is more likely with internalizing symptoms sinc e there is a degree of ambiguity and symptoms are not outwardly evident. Along th is same line, the social attribution
8 theory states that ambiguous stimuli are more liable to distortion than more obvious or observable stimuli. Inte rnalizing symptoms are seen as more ambiguous since the symptomatology take s place more within a person and therefore is less resistant to distortion as induced by parental psychopathology. Externalizing symptoms are more r eadily observable and therefore more resistant to distortion due to parental psychopathology. Other studies found less evidence for the influence of parental psychological symptoms. Bingham, Loukas, Fitzgerald, and Zucker (2003) studied families with alcoholism and assess ed parental ratings of child behaviors using the Child Behavior Checklist (CBC L; Achenbach, 1991). The parental level of functioning due to alcoholism did not im pede accurate child behavior ratings. ParentsÂ’ ratings corresponded to the theoretic al structures of their sonÂ’s behavior problems and accurately reflected the behav iors that each parent experienced. Although parental agreement was still lo w, mothersÂ’ and fathersÂ’ ratings accurately indicated differences in t he behaviors that they witnessed. This pattern again points to the importance of obtaining multiple informants and including both parents in the assessm ent and research of childrenÂ’s behavior problems. The authors of this article argued that at tention should be paid to individual cases involving excessive im pairment in cognitions and perceptions due to long-term alcohol use. Limitations of this study warrant future research because the sample was primarily white and the children were all male. Also, the children were young (aged 3-5) and theref ore, it might not be possible to generalize these findings to older children an d adolescents. In short, the effects
9 of parental psychological symptoms on the ratings of child behavior problems are apparent. The theories under lying these effects, su ch as the distortion hypothesis and the social attribution t heory were considered for the present study. Interparental Conflict and Family Distress Marital discord and overall family di stress have also been indicated as factors that could influ ence parental agreement on child renÂ’s behavior problems. Family distress is particularly heightened when parents must tend to a child with severe emotional and behavioral problem s. This process can lead to disagreements between parents on how the child should be cared for and can place an additional strain on a marital relationship. Christensen et al. (1992) studied three kinds of families: thos e with marital discord and child conduct problems, those with either marital discord or child conduct problems, and those with neither marital discord nor child c onduct problems. Pa rental disagreement increased as levels of family distress in creased, such that those families with neither marital problems nor child behav ior problems had the lowest level of disagreement between parents. Families with one sour ce of distress (either marital discord or child behavior problem s) had less parental disagreement than families with both sources of distress, but were still higher on disagreement than those families with no distress. Another study by Webster-Stratton (1989) compared maritally supported families, maritally distressed families, and single parent families on parental perceptions of child adjustment, ch ild behavior problems, and parenting
10 behaviors. Single parent families only included single mothersÂ’, not single fathersÂ’, reports of childrenÂ’s behavior problems. Overall, single mothers reported more total child behavior probl ems and higher stress than maritally supported mothers. Both single and mari tally distressed mothers reported more stress than maritally supported mothers, with single mothers reporting the most stress. It may be that si ngle mothers were more negative in reports of childrenÂ’s behavior and their own parenting behaviors than maritally distressed mothers because single mothers are intera cting more with the child and overcompensating for the la ck of another parent whereas the maritally distressed mothers are not necessarily doing the work for both parents. A major limitation to Webster-StrattonÂ’s (1989) study is t he lack of analyses of mothersÂ’ versus fathersÂ’ ratings of child behavior pr oblems and the overwhelming focus on mothersÂ’ perceptions. The current study focuses on differences between mothers and fathers and will analyze behavior ratings for bot h. Frosch and M angelsdorf (2001) found no relationship between mothersÂ’ and father sÂ’ reports of child behavior problems and observed marital behaviors. However, there was an association between observed marital problems and observersÂ’ ra tings of child behavior problems. The lack of association between parental ratings and marital problems was due to the relatively high level of functioning that the parents repor ted. Overall, the association between marital discord, fa mily distress, and parental agreement on child behavior problems have been neglected in research (Duhig et al., 2000). The present study will further analyze these relationships.
11 Types of Child Behavior Discrepancies in parental ratings of behavior problems can be examined for the types of child behaviors. Achenbac h et al. (1987) discussed two types of behavior problems: overcontrolled vers us undercontrolled. Overcontrolled problems are also referred to as inter nalizing behaviors and include designations such as, withdrawn, anxious, depressed, psychosomatic, and fearful. On the other hand, undercontrolled or externa lizing behaviors are described as, antisocial, aggressive, hyperactive, assaul tive, and sociopathic. Individual items assessing internalizing or externaliz ing behavior problems differ in their agreement across parents. Duhig et al. (2000) found that mothers and fathers exhibited fewer differences in their ra tings of externaliz ing behavior problems than ratings of internalizing or total behav ior problems. However, the differences across this meta-analysis were small and n onsignificant. As mentioned earlier, the social attribution theory (Dodge, 1986) states that more internalizing or ambiguous stimuli would be more prone to distortion and therefore lower agreement across parents than externalizing or more readily observable stimuli. Researchers have also suggested that internalizing child behaviors are less stable across situations than externaliz ing behaviors (Stanger & Lewis, 1993). An example of this scenario could be a child, whose mother displays psychological symptoms, may be more likely to show internalizing symptoms when at home and around the mother versus at school. Likewise, if internalizing behaviors are less stable, then the child may display different symptoms when around the mother versus the father.
12 In addition, parental agreement tends to differ depending on the age of the child. Achenbach et al. (1987) found that parents had more consistent agreement for younger children aged 6 to 11 years old and for externalizing problems than for adolescents and internaliz ing problems. Lastly, Christensen et al. (1992) studied specific items of the Child Behavior Checklist (CBCL; Achenbach, 1991) and analyzed the associat ion of item characteristics on agreement. Items rated for high objectivity, observability, social undesirability, and disturbance evidenced lower discrepancies in parental ratings than those lower on these characteristics. Since it ems on the externalizing scales of the CBCL tend to be higher on observability objectivity, and social undesirability than on the internalizing scales, thes e items show higher interparental agreement. Gender Differences Inherent differences in how males ve rsus females view behavior problems could exist regardless of parental status Symptom perception differences have been found between men and women (Macintyr e, 1993). MenÂ’s self report of common cold symptoms were more seve re than those rated by clinicians, whereas womenÂ‘s reports had greater co rrespondence with cliniciansÂ’ reports. Therefore, men appeared to exaggerate symptoms more often than women. However, there are differences betw een perceptions and reporting. For example, women may more often t han men report a symptom once it is perceived (Mechanic, 1976). However, this difference gets smaller when objective measures of sym ptoms are used, w hen symptoms are more tangible or
13 observable, and when symptoms are more severe (Mechanic, 1976). This finding supports the social attribution theory regarding the gr eater agreement of parents on externalizing rather than internalizing behaviors in children. There also could be differences in how mothers versus fathers view their daughtersÂ’ versus their sonsÂ’ behavior and their reactions to such behavior may also differ. Socialization differences we re evident in a study that investigated mothersÂ’ reactions to videotapes of childr en engaging in injury-risk activities on a playground (Morrongiello & Dawber, 2000) Mothers of daughters were more likely to rate behaviors as posing a high degree of injury risk and they intervened more quickly than mothers of sons. Also the speed to intervene was positively associated with their childÂ’s in jury history, in that children with many injuries in the past had mothers who took longer to intervene in risk taking behaviors. MothersÂ’ verbalizations to childrenÂ’s risk taking were also evaluated. Mothers of daughters gave more cautionary st atements and communicated more vulnerability about potential injury w hereas mothers of sons gave more statements encouraging risk taki ng behaviors. This study did not look at fathersÂ’ reactions to risk taking behaviors. It woul d be interesting to see if there would be a higher or similar degree of encouragem ent to boys by fathers than mothers given the tendency of parents in general to promote aggre ssive behaviors in their sons (Morrongiello & Dawber, 2000). Perh aps, fathers and mothers would both underreport dangerous or externalizing beh aviors problems in boys and show more agreement given that they might have similar views on how boys should behave.
14 Gender role socialization may account for differences in how mothers versus fathers react to behav iors in their sons versus their daughters. Women are often concerned with relationship s and acceptance (Timmers, Fischer, & Manstead, 1998). They may express emot ions that strengthen relationships, such as sadness or empathy, and inhibit emotions that coul d be detrimental to relationships, such as anger or aggressiv eness. Men, on t he other hand, are more likely to express emotions of contro l and pride and are less likely to express emotions that make them mo re vulnerable (Timmers et al., 1998). Therefore, men could view a childÂ’s externalizing behavior as less negative given that they value expressions of power and anger. Wo men might see a childÂ’s internalizing behavior as less negative given that they value more vulnerable expressions of sadness and empathy. Gender and socializ ation differences offer alternative explanations for why parental disagr eement on child behavior problems could exist. The Present Study All of these issues relate to how mothers and fathers view childrenÂ’s behavior. More specifically, do mothers and fathers differ in how they rate childrenÂ’s internalizing and externalizing behaviors and do these differences depend upon the level of their own psyc hological symptoms, levels of interparental conflict, or the characteri stics of the behaviors themselves? Because clinicians rely heavily upon par ental ratings of child behaviors in assessing child psychopathology, this study attempted to identify those factors
15 that lead to the greatest disagreem ent between parents on child behavior problems. The present study differed from previous research in that it controlled for the amount of contact betw een parents and the children they were rating. By taking one factor, amount of contact with the target child, out of the equation, it is possible to examine whether differences in ratings are accounted for by other factors. However, amount of general contact that these parents have with children for instance, in their home or pr ofession was still evaluated. In contrast to previous studies that have looked at pai rs of parentsÂ’ ratings of their own child, this study was conducted on mothers and fathers who rated a videotape of a child whom they did not know acting in ei ther an internalizing, externalizing, or non-clinical manner. By controlling for the amount of cont act with and actual knowledge of the child, it was possible to examine if other factors, such as parental psychological symptoms, interparental conflict, and ty pe of behavior, were responsible for mothersÂ’ and fathersÂ’ disagreement in ra ting behavior. In addition, by using an experimental design with direct observati on, errors due to retrospective ratings from memory were elimi nated. The social attrib ution theory states that ambiguous environmental stimuli are more liable to infer and distort perceptions than more obvious stimuli (D odge, 1986). Therefore, the first hypothesis stated that there would be smalle r differences in ratings between mothers and fathers on externalizing behaviors of children than internalizing behaviors because externalizing behaviors are more observable and more re sistant to distortion.
16 The second hypothesis stated that ther e would be a significant interaction between parent gender and type of behavior vi ewed, in that mothers would rate internalizing behavior problem s higher than fathers. Ho wever, this difference was not expected for externalizing behaviors. The distortion model states that parental psychologic al symptoms can inflate or distort ratings of child behavior problems (Richters, 1992). Therefore, the third hypothesis stated t hat higher levels of parent al psychological symptoms and higher levels of interparental conflict in mothers and fathers would be related to higher ratings of childrenÂ’s behavio r problems in the videos. In addition, because parents with more psychological symptoms tend to have children with greater psychopathology (Connell & Goodm an, 2002; Kane & Garber, 2004), the fourth hypothesis stated that higher ra tings of a parentÂ’s own childÂ’s behavior problems would be related to higher ratings of behavior problems of the child in the video.
17 Method Participants A total of 79 mothers and 71 fathers we re recruited to participate in the study. A power analysis (with a power of .80, alpha set at 05, and expecting a medium effect size) showed that a mi nimum of 64 mothers and 64 fathers were needed to test for mean differences adequately (Cohen, 1992). Thus, the sample size should be adequate to test fo r main effects. However, a post-hoc power analysis revealed that the sample size may not have been adequate to test for an interaction effect. Specif ically, in order to test adequately for interaction effects a total of 81 mother s and 81 fathers would have been required. Parents all had at least one child between the ages of 4 and 21. Mothers ranged in age from 18 to 56 years (M =42.71, SD =7.19) and fathers ranged in age from 25 to 58 years (M =45.35, SD =6.44). Regarding race and ethnicity, the sample was primarily Caucasian (mothers 90.9%, fathers 93.0%), with some parents of African American (mothers 7.8%, father s 2.8%), Hispanic (mothers 0%, fathers 2.8%), and Asian (mothers 1.3% fathers 1.4%) ethnicities The majority of the sample was married (mothers 86.1%, fat hers 95.8%), while the remainder were separated (mothers 1.3%, fathers 1.4%), divorced (mothers 5.1%), or single (mothers 7.6%, fathers 2.8%). Also, t he majority of the sample was employed (mothers 77.2%, fathers 95.7%) while the remainder were unemployed (mothers 6.3%, fathers 1.4%), reti red (mothers 1.3%) or other (mothers 15.2%, fathers
18 2.9%), which included being a student. MothersÂ’ mean socioeconomic status (SES, Hollingshead, 1975) was 50.73, and fathersÂ’ mean SES was 53.01. Thus the sample showed relatively high SES. A total of 32.9% of mothers reported that either themselves or their childÂ’s fa ther had received mental health services in the present or past, while 67.1% said that they had not. A total of 29.6% of fathers reported that either themselves or their childÂ’s mother had received mental health services, while 70.4% reporte d they had not. A total of 26.6% of mothers and 16.9% of fat hers reported that at leas t one of their children had received mental health services. See T able 1 for other mothersÂ’ and fathersÂ’ demographics. Overall, mothers and fathers did not differ significantly on socioeconomic status, the number of child ren they had living in their home, whether they had daughters only, sons only, or both sons and daughters, or how many siblings they had growing up (all pÂ’s > .05). In addition, they did not differ on the gender and age of their child closest in age to 6 and the total behavior score of this child on the CBCL (all pÂ’s > .05). They did, however, differ on marital status and age in which more mothers (14%) than fathers (4%; 2 (1, N=150)=4.16, p=.04) were not married and mothers (M =42.71 years old) tended to be younger than fathers (M =45.35 years old; t (148)=-2.36; p =.02). This difference could be due to the fact that single mothers t end to be younger on average.
19 Table 1. MothersÂ’ and fat hersÂ’ demographic variables. Variable Mothers (N=79) Fathers (N=71) Mean age 42.71 (7.19) 45.35 (6.44) Mean number of children 2.53 (1.08) 2.49 (0.98) Mean percentage of professional involvement with children 24.31 (38.34) 8.62 (22.70) Mean experience with children 7.59 (2.52) 6.30 (2.43) Mean weekday time 5.59 (3.21) 4.41 (3.70) Mean weekend time 10.20 (3.54) 8.15 (4.48) Mean SES 50.73 (8.88) 53.01 (9.11) Race/Ethnicity (%) Caucasian African American Hispanic Asian 90.9 7.8 0.0 1.3 93.0 2.8 2.8 1.4 Marital Status (%) Married Separated Single, with partner Divorced Single, no partner 86.1 1.3 5.1 5.1 2.5 95.8 1.4 1.4 0.0 1.4 Gender child (%) Male Female 59.7 40.3 57.7 42.3 Type of children (%) Sons only Daughters only Both 27.3 13.0 59.7 27.1 15.7 57.1 Note: Standard deviations are in parentheses.
20 Measures Interview/Videotape Stimulus. Videotapes were developed by the researchers and included an 8 year old boy and an 8 year old girl child actor being interviewed for five minutes usi ng questions from the Semistructured Clinical Interview for Children and Adolescents (SCICA, McConaughy & Achenbach, 2001; Appendix A). The age of eight was chosen because children can still show visible manifestations of their behavior (e.g. throwing a toy, turning away from the interviewer and sulking) but can also verbalize their own experiences (e.g. reporting anger or sadne ss). The SCICA is a protocol of questions pertaining to a childÂ’s school, activities, friends, family relations, fantasies, self perceptions, and parent/t eacher problems. The SCICA can be used with children from ages 6 to 18. Mean test-retest reliability over a 12-day period was .78 across empiricall y based syndrome scales and broad DSMoriented scales. For 6-11-year-olds, inte rnal consistency reliability ranged from .61-.88 across empirically based syndrom e scales and from .58-.74 across broad DSM-oriented scales. The SCICA was c hosen so that extensive information could be conveyed through the videotaped obse rvations-both from what the child says and how the child behaves. Child Behavior. The participants rated the videotape using 4 subscales from the Child Behavior Checklist (CBCL, Achenbach & Rescorla, 2001; Appendix B). The CBCL is a parent-report measure of child behavior problems for children ranging in age from 6-18 years old. The four subscales that were used were the Anxious/Depressed s ubscale, the Withdrawn/Depressed
21 subscale, the Aggressive Behavior sub scale, and the Rule Breaking Behavior subscale. These four subscales were chosen to provide a thorough assessment of internalizing (anxious-depressed, withdrawn-depressed) and externalizing (aggressive behavior, rule-breaking behavior) symptoms. Participants also rated their own child on the entire CBCL (not included in appendix due to copyright issues). If the parent had more than one child, then the child closest in age to 6 was rat ed. The age of six was chosen because children can show visible manifestations of their behavior, can verbalize their own experiences, and are similar in age and development to the child in the video. If the child closest in age to 6 was younger than 6, then the preschool version of the CBCL (Child Behavior Che cklist for Ages 1 to 5) was used. If the child closest in age to 6 was older t han 18 then the adult version of the CBCL (Adult Behavior Checklist for Ages 18 to 59) was used. All of these measures lead to T-scores of internalizing, exte rnalizing, and total behavior problems. Higher scores indicate higher child behav ior problems. The CBCL has good psychometric properties. The internal cons istency reliability ranges from .80-.94 for the broadband Internalizing, Externaliz ing, and Total Behavior Problems and the test-retest reliability over a two-week period ranges from 82-.91 for the four subscales used for rating the child in the video (Achenbach & Rescorla, 2001). In the current sample, internal co nsistencies ranged from .93-.94. Parental Psychological Symptoms. In addition, partici pants were asked to report on their own psychological symptom s using the Brief Symptom Inventory (BSI, Derogatis, 1993; not included in appe ndix due to copyright issues). The
22 BSI is a self-report measure used to ident ify clinically rele vant psychological symptoms in adolescents and adults. It contains 53 items covering nine symptom dimensions: Somatization, Obsession-Compulsion, Interpersonal Sensitivity, Depression, Anxi ety, Hostility, Phobic Anxiet y, Paranoid Ideation, and Psychoticism; and three global indices of di stress: Global Severi ty Index, Positive Symptom Distress Index, and Positive Symp tom Total. These indices measure current level of symptomatology, intens ity of symptoms, and number of reported symptoms, respectively. Good internal c onsistency (.71-.85) is reported for the nine dimensions. Test-retest reliability for the nine di mensions ranged from .68.91 and test-retest reliability for the three Global Indices ranged fr om .87-.90. In the current sample, the internal consist ency was .92. The Gl obal Severity Index was used for this study. Higher scores on the BSI indicate greater psychological symptoms. Interparental Conflict. Participants were also a sked to report on the levels of interparental conflict in their fam ily using the OÂ’Leary Porter Scale (OPS, Porter & OÂ’Leary, 1980; Appendix C). This measure is a 10-it em parent-report measure on the frequency of various forms of marital hostility, including quarrels, sarcasm, and physical abuse, that take pl ace in front of the child. Higher scores on the OPS indicate greater interparental conflict. Internal consistency reliability was .86 and test-retest reliability over a two-week period wa s .96. One item about father/husbandÂ’s role in the fam ily (item #11) has been added to balance out the inclusion of item #4 about mother s. Internal consistency remains high
23 with the addition of this item (Epstein, Renk, Duhig, Bosco, & Phares, 2004). In the current sample, the inter nal consistency was .80. Demographics and Time Spent. Participants were also asked to fill out a basic demographic questionnair e including questions about the amount of time spent with their own child(ren), number of siblings in their childhood family, professional involvement, and general amou nt of contact with children (Appendix D). Procedures Development of Videos and Pilot Study. The development of the videos began with the selection of two child ac tors, one male and one female, both around the age of 8. The boy child actor wa s the brother of a research assistant in the research group. The mother and actor signed a brief informed consent before coming in to tape the video. The boy child actor, his mother, and his sister came into the lab for the training session for the development of the video. He was instructed to think of a time when he was really sad and upset in order to act in an internalizing manner. He then prac ticed answering the questions from the SCICA (Appendix A) acting in this manner. Next, he was instructed to think of a child who is very active, Â“bouncing off the wallsÂ”, and who canÂ’t sit still, in order to act in an externalizing manner. He then practiced answering the same questions acting in this manner. Finally, he was in structed to answer the questions acting as he normally would, neither upset, nor too active. After each set of questions were asked, the principal investigat or and the child actor discussed how he would act in a free play situati on in each of these manners.
24 The child actor and the principal invest igator (PI) then taped a rehearsal of the PI interviewing the boy child actor and three minutes of free play behavior while acting in each of three manners. The taping was held in one of the child rooms in the university psychological clinic The room contained a long table in which the boy child actor sat in view of the camera and the PI sat behind the camera. The room also contained seve ral toys, including toy cars, board games, and coloring book for the child actor to engage with during free play. A week later, the child actor and the PI taped the final videos. Two different segments of the externalizing free play were taped. One was thought to be more intense than the other. The more int ense segment was used in the initial pilot study. The girl child actor was selected due to her similarity in appearance and age of the boy child actor. The gi rl child actor was the daughter of a departmental faculty member. Agai n, the parent and daughter signed an informed consent form before coming in to tape the video. The girl child actor and the PI practiced each of the segments in the same manner as the boy child actor. In addition, the gi rl watched the segments of the boy child actor in order to match her behaviors and intensities of behaviors to his. Later, the girl child actor and the PI taped the final video segments of the interview and free play in each of the three manners. Both child actors were given gift certificates for their participation. The pilot study began after several copies of the videos of the boy child actor segments and girl child actor segments were made. Ten upper level doctoral clinical psychology graduate students who had clinical and research
25 experience with children were recruited. Th e participants (N=10) were all female. Nine were Caucasian (90%) and one wa s African American (10%). Six participants already had their MasterÂ’s Degree and four were currently working on their MasterÂ’s thesis. They had a mean of 1.75 years (SD=1.25) working with child clients and a mean of 3.40 years (S D=2.58) of professi onal experience with children. They had taken a mean of 2.70 (SD=1.89) classes in child development and psychopathology. Each participant was given a copy of the video that contained all six segments (boy -internalizing, boy-externalizing, boynonclinical, girl-internalizing, girl-externalizing, girl-n onclinical) and a rating form (Appendix E) that asked to rate whether the segment showed internalizing, externalizing, or non-clinic al behavior and to rate the intensity of the behavior in each segment. The rating form comprised of a forced-choice design in which the participants could only rate one boy child actor segment internalizing, one externalizing, and one non-c linical. Each category needed to be used once and only once. The same rules applied to the girlsÂ’ segments. Based on the pilot study results, there was 100% agreement across all ten participants on whether the video segments showed internalizing, exte rnalizing, or non-clinical behavior. Using a dependent t-test, the intensities between the participantsÂ’ ratings of the girl-internalizing and boy-i nternalizing, girl-externa lizing and boy-ext ernalizing, and girl-nonclinical and boy-nonclinical behaviors were compared. Nonsignificant results between boy and girl in tensities were expected. There were non-significant differences between the in tensity ratings of the girl and boy internalizing segments (t (9)=-.612, p =.555) and the intensity ratings of the girl
26 and boy non-clinical segments (t (9)=1.00, p =.343). However, there were significant differences between the in tensity ratings of the girl and boy externalizing segments (t (9)=-6.00, p <.01), with higher ratings for the boy. Because there were significant results between the girl and boy externalizing intensity ra tings, the video for the boy acting in an externalizing manner was modified and re-piloted. Two mi nutes of the less intense free play of the boy externalizing behavior and one minute of the more intense free play were combined, instead of the full three minutes of the more intense free play that was initially piloted. Using the same ten participants, a video with the boy externalizing interview and free play segm ent was distributed and the participants were again asked to rate whether t he segment showed internalizing, externalizing, or non-clinic al behavior and to rate the intensity of the behavior on an additional rating form (Appendix F). Again, there was 100% agreement across participants that the boy segment showed externalizing behavior. There also were non-significant results betw een the girl-externalizing segment and the modified boy-externalizing segment (t (9)=-.429, p=.678). Rather than have the participants rate the video wit hin-subjects, perhaps, t he participants should have rated the videos between-subjects in order to correspond more closely to the design of the actual study. In short, the pilot study was successful and actual data collection using the m odified videos began. Present Study. Mothers and fathers were recr uited via flyers throughout the campus of a southeastern univers ity and around the community, through an online participant pool through the Psycholog y Department at the university, via
27 letters that were sent out to researchersÂ’ friends and family, and through word of mouth. Participants who were student s in the Psychology Department were given extra credit points towards one of t heir psychology courses for partaking in the study. If participants we re not students, then they were entered into a raffle to receive one of two $50 cash prizes, or a $100 cash prize. To meet eligibility, the participant needed to be a parent of a child. A Â“parentÂ” was defined as an individual who has at least monthly face -to-face contact with the child. Thus, biological parents, stepparents, and adoptive parents were all included if they had sufficient contact with their child. It is important to not e that parentsÂ’ inclusion was based on the age of their own child. Parents were screened to ensure that they had at least one child who was between 4 and 21 years old. Once mothers and fathers were recr uited and had met the criteria to participate (i.e., they must be a parent of at least one child between the ages of 4 and 21 and must have at least monthly face -to-face contact with the child) they were asked to sign the informed cons ent form (Appendix G) and then were assigned to view a videotape. Using a between subjects, experimental design, they were randomly assigned to one of 6 condi tions that reflect the type of video: girl-internalizing, girl-ext ernalizing, girl-nonclinical, boy-internalizing, boyexternalizing, and boy-nonclinical. After vi ewing the video, they rated the child in the video using the 4 subscales from t he CBCL. After they completed the behavior ratings, they were asked to co mplete the demographi c questionnaire, the BSI, and the OPS. Additionally, they we re asked to rate their own child using the CBCL. If they had more than one child, they were asked to rate the child that
28 is closest in age to 6 years old but al so still within the 4-21 age range. These measures were given after the videotape and after completion of the behavior ratings in order to reduc e demand characteristics on their behavioral ratings of the child actor on the CBCL. Partici pants viewed the videotapes either alone or in small groups (with instructions to rema in silent during viewi ng) and if in small groups they completed their measures i ndividually in the sa me room. At the completion of the study, the participants were given assigned extra credit points for their participation in t he study or gave their name, address, email, and phone number on a separate sheet of paper from their materials to be entered into the raffle. They were given a debriefi ng form (Appendix H), and thanked for their participation. Because of some difficulty in recr uiting participants to participate in person, some participants participated by hav ing the materials sent to them by mail. One mother (1.3%) and six fathers (8.5%) were mailed packets containing detailed instructions, a DVD of one of t he 6 segments, the informed consent, the 56 behavioral items from the CBCL, the demographic form, the BSI, the OPS, the CBCL to rate their own child, and a debrie fing form. In additi on, an index card was included for the participant to write their name, address, email, and phone number so that they could be entered into the raffle. A postage paid envelope was also enclosed for the participant to send back all completed materials. Participants who completed the study in person versus those who completed the study via mail did not di ffer on many variables including age, socioeconomic status, number of childr en, number of siblings growing up,
29 percentage of professional involvement spent with children, amount of experience with children other than t heir own, their childÂ’s age and gender closest to six, their childÂ’s total behavio r problems, their marital status, and the type of children they have (all pÂ’s > .05) There was a significant difference between the number of mot hers and the number of fathers recruited by mail because fathers were more difficult to recruit in person and therefore, the researchers were able to re cruit more fathers by allo wing them to participate on their own time in their own home (p<.05). At least 10 mothers and 10 fathers were randomly assigned, viewed, and rated each video. A total of 11 mother s (13.9%) and 11 fathers (15.5%) viewed the girl-internalizing video, 13 mothers (16.5%) and 10 fathers (14.1%) viewed the girl-externalizing video, 10 mothers (12.7%) and 11 fathers (15.5%) viewed the girl-nonclinical video, 17 mothers (21.5%) and 16 fathers (22.5%) viewed the boy-internalizing video, 16 mothers (20.3%) and 13 fat hers (18.3%) viewed the boy-externalizing video and, 12 mothers (15.2%) and 10 fathers (14.1%) viewed the boy-nonclinical video.
30 Results Randomization and Reliability Checks ANOVAs and Chi-square tests were performed in order to verify that random assignment was successful in equa lizing parental characteristics across the groups. It was determined that t he parentsÂ’ gender, age, marital status, socioeconomic status, number of children they had, number of siblings they had, percentage of professional involvement spent working with ch ildren, amount of experience they had with children other t han their own, the gender of their child closest in age to 6, and their childÂ’s total behavior problems (total T-score on CBCL) did not differ depending on which video they rated (all pÂ’s > .05; See Table 2). However, the age of participant sÂ’ child closest to 6 did differ between videos (p<.01). Therefore, follow up analyses were performed and Pearson correlations showed that the childÂ’s age clos est to 6 was not related to any of the dependent variables (the four subscale rati ngs of the videos, the internalizing mean, the externalizing m ean, and the total mean). In addition, the parentsÂ’ externalizing ratings of their own child on the CBCL did differ depending on which video they saw, such that those who sa w the boy-nonclinical video had a higher externalizing T-scores for their own ch ild versus those who saw the girlexternalizing video (p<.05). This result could be the case of randomization not being completely effective. It also c ould be because the parent rated their own child after viewing the video and the boynonclinical video primed the parent to
31 view their own child more severely. This question should be followed up in future research. Overall, randomization acro ss videos appeared to be effective other than these minor differences which do not appear to be related to the dependent variables in the study. Table 2. Analysis of variance and chisquare tests to determine whether random assignment across videos was effective. Variable Total df F p Age 149 .99 .43 SES 138 2.21 .06 Number children 149 .77 .57 Number siblings 149 1.40 .23 Professional involvement 142 .63 .68 Experience with children 148 .08 .99 Age of child 147 4.19 .00** Internalizing behavior 147 .40 .85 Externalizing behavior 147 2.47 .04* Total behavior 147 1.93 .09 Variable Df 2 p Parent gender 5 .54 .99 Marital status 5 7.73 .17 Gender of child 5 10.08 .07 p<.05, **p<.01 Differences Between Mothers and Fathers on Video Ratings Descriptive statistics for mother sÂ’ and fathersÂ’ mean subscale, internalizing, and externalizi ng ratings of the videos are provided in Table 3. To test the first two hypotheses, a series of four 3x2x2 multivariate analyses of variance (MANOVAs) were used because t he dependent variables, the scores of the four subscales of the CBCL, show multicollinearity with each other. The factors include: type of video (internalizi ng versus externalizing versus non
32 Table 3. MothersÂ’ and fathersÂ’ descr iptive statistics of their mean ratings averaged across the six videos. Mothers Fathers Mean SD Mean SD Girl Internalizing Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.45 1.32 0.16 0.29 0.78 0.22 0.29 0.42 0.17 0.33 0.26 0.24 0.47 1.56 0.21 0.29 0.88 0.25 0.53 0.37 0.28 0.35 0.40 0.30 Girl Externalizing Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.57 0.33 0.27 0.74 0.48 0.51 0.30 0.23 0.14 0.32 0.25 0.21 0.63 0.42 0.38 0.89 0.54 0.65 0.24 0.26 0.44 0.54 0.22 0.46 Girl Non-Clinical Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.32 0.16 0.19 0.12 0.26 0.16 0.12 0.21 0.16 0.13 0.14 0.14 0.31 0.10 0.18 0.23 0.23 0.21 0.25 0.21 0.22 0.41 0.23 0.31 Boy Internalizing Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.95 1.22 0.15 0.19 1.05 0.17 0.57 0.52 0.18 0.19 0.51 0.16 0.71 1.08 0.08 0.21 0.85 0.15 0.39 0.43 0.12 0.24 0.38 0.17 Boy Externalizing Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.42 0.37 0.50 1.05 0.40 0.78 0.35 0.27 0.42 0.52 0.30 0.44 0.44 0.36 0.38 1.04 0.41 0.72 0.37 0.23 0.20 0.46 0.28 0.32 Boy Non-Clinical Anxious-Dep Withdrawn-Dep Rule-Breaking Aggressive-Beh Internalizing Externalizing 0.27 0.06 0.13 0.26 0.19 0.20 0.13 0.10 0.07 0.25 0.10 0.15 0.70 0.33 0.34 0.79 0.56 0.57 0.39 0.35 0.25 0.59 0.30 0.41 clinical behaviors), parent gender (mother versus father), and child gender (boy child actor versus girl child actor).
33 The first hypothesis stated that th ere would be smaller differences in ratings between mothers and fathers on exte rnalizing behaviors of children than internalizing behaviors because externaliz ing behaviors are more observable and more resistant to distortion. Therefor e, it was expected t hat there would be a significant difference between mothersÂ’ and fathersÂ’ ratings of the internalizing behavior videos but there would not be a significant difference for the externalizing behavior videos. The sec ond hypothesis stated that there would be a significant interaction between par ent gender and type of behavior in the video whereby, mothers would have signi ficantly higher ratings on the CBCL subscales for the internalizing behavio r videos compared with the fathers but there would not be a significant differ ence between mothers and fathers on their ratings for the externaliz ing behavior videos. Althoug h the overall MANOVA was significant (F (4, 135)=141.35, p <.001), results showed that there was no main effect for parent gender (F (4, 135)=1.50, p =.21) nor was there a significant interaction for parent gender and type of video (F (8, 270)=1.04, p =.41). See Table 4. Thus, hypothesis 1 was not suppor ted in that there were no differences between mothers and fathers on their rati ngs of the internalizing videos. Likewise, hypothesis 2 was not supported gi ven that there was not a significant interaction between parent gender and type of video in which mothers rated internalizing videos higher than fathers.
34 Table 4. Multivariate and univariate F va lues for Parent Gender by Type of Video by Child Gender of Video interactions for pa rentsÂ’ ratings of the video on the four subscales of the CBCL. Multivariate Univariate Variable All Anxious Depressed Withdrawn Depressed Rule Breaking Aggressive Behavior Parent Gender (P) 1.50 0.68 1.31 0.56 4.55* Type of Video (T) 55.23*** 5.15** 163.51*** 12.52*** 48.16*** Child Gender of Video (C) 5.47*** 4.00* 2.06 0.45 6.68* P by T 1.04 2.32 0.11 0.65 2.16 P by C 0.99 0.12 0.24 0.32 .54 T by C 8.25*** 7.14** 3.82* 1.79 4.40* P by T by C 1.38 2. 71 3.33* 2.68 1.81 *p<.05 **p<.01 ***p<.001 The following results are summarized from the MANOVAs but were not part of the hypothesis testing. There was a significant ma in effect for the type of video (F (8, 270)=55.23, p <.001) and the childÂ’s gender in the video (F (4, 135)=5.47, p <.001). Follow-up univariate tests showed that all subscale mean scores were significant for the ty pe of video: Anxious-Depressed (F (2,138)=5.15, p <.01), Withdrawn-Depressed (F (2,138)=163.51, p <.001), Rule-Breaking (F (2,138)=12.52, p <.001), and Aggressive Behavior (F (2,138)=48.16, p <.001). The Anxious-Depressed subscale mean (F (1,138)=4.00, p <.05) and the Aggressive Behavior subscale mean (F (1,138)=6.68, p <.05) were found to be
35 significant for the childÂ’s gender in the video. See Table 4. Tukey post hoc follow-up tests for type of video s howed that on the Anxious-Depressed subscale, the internalizing videos (M =0.64, SE =0.05) differed significantly from the non-clinical videos (M =0.40, SE =0.06) and the externalizing videos (M =0.52, SE =0.05) but the non-clinical videos and t he externalizing videos did not differ significantly. Thus, the internalizing vi deos were rated as showing more anxiousdepressed symptoms than the other two videos. On the Withdrawn-Depressed sub scale, the nonclinical videos (M =0.16, SE =0.05), the externalizing videos (M= 0.37, SE=0.05) and t he internalizing videos (M =1.29, SE =0.05) all differed significantly from each other. Again, the internalizing videos were rated as showing the most withdrawn-depressed symptoms overall and the ex ternalizing videos were rated as showing more withdrawn-depressed symptoms t han the non-clinical videos. On the Rule-Breaking subscale, the internalizing videos (M =0.15, SE =0.03) and the nonclinical videos (M =0.21, SE =0.04) differed significantly from the externalizing videos (M =0.38, SE =0.03) but the internalizing videos did not differ significantly from the nonclincal vi deos. This finding suggests that the externalizing videos were rated as sho wing more rule-breaking symptoms than the other videos. Lastly, on the Aggressive Behavior s ubscale, the internalizing videos (M =0.24, SE =0.05) and the nonclinical videos (M =0.35, SE =0.06) differed significantly from the externalizing videos (M =0.93, SE =0.05) but the internalizing videos did not differ significantly from the nonclincal videos. Thus, the
36 externalizing videos were rated as sho wing more aggressive behavior symptoms than the other videos. Tukey post hoc follow-up tests for child gender show that on the AnxiousDepressed subscale, the boy child actor videos (M =0.58, SE =0.04) were rated significantly higher than girl child actor videos (M =.46, SE =0.05). On the Aggressive Behavior subscale, again the boy child actor videos (M =0.59, SE =0.04) were rated significantly higher than the girl child actor videos (M =0.43, SE =0.05). Therefore, the boyÂ’s behavior was seen as more extreme overall. A significant interaction between the type of video and the childÂ’s gender was found (F (8,270)=8.25, p <.001). Univariate follow-up tests showed that the interaction was significant for t he Anxious-Depressed subscale mean (F (2,138)=7.14, p <.01), the Withdrawn-D epressed subscale mean (F (2,138)=3.82, p <.05), and the Aggressive Behavior subscale mean (F (2,138)=4.40, p <.05). Tukey post hoc tests showed that the participantsÂ’ ratings on the Anxious-Depressed sub scale for the boy-internalizing video (M =.83, SE =.07) were significantly higher than the girl-nonclinical video (M =.315, SE =.08), the boy-externalizing video (M =.43, SE =.07), the girl-internalizing video (M =.46, SE =.08), and the boy-non clincal video (M =.47, SE =.08). However, ratings on the boy-internalizing video did not differ significantly from the ratings on the girl-externalizing video (M =.60, SE =.08). See Table 5 and Figure 1. Thus, participants rated the boy-interna lizing video as showing more anxiousdepressed symptoms than the girl-internalizing video but similar amounts of symptoms to the girl -externalizing video.
37 Table 5. Tukey post hoc tests for the Ty pe of Video by Child Gender interaction across the Anxious-Depressed, Withdrawn-Depressed, and Aggressive Behavior subscale scores. Anxious-Depressed Mean Video 1 2 3 4 Girl-NC .32 Boy-Ext .43 Girl-Int .46 Boy-NC .47 Girl-Ext .60 .60 Boy-Int .83 Withdrawn-Depressed Mean Video 1 2 3 4 Girl-NC .13 Boy-NC .18 Boy-Ext .36 Girl-Ext .37 Boy-Int 1.15 Girl-Int 1.44 Aggressive-Behavior Mean Video 1 2 3 4 Girl-NC .18 Boy-Int .20 .20 Girl-Int .29 .29 Boy-NC .50 .50 Girl-Ext .81 .81 Boy-Ext 1.05 Tukey post hoc tests also showed that participantsÂ’ ratings on the Withdrawn-Depressed subscale fo r the girl-internalizing video (M =1.44, SE =.07) were significantly higher t han the girl-nonclinical video (M =.13, SE =.07), the boynonclinical video (M =.18, SE =.07), the boy-externalizing video (M =.36, SE =.06), the girl-externalizing video (M =.37, SE =.07), and the boy-internalizing video (M =1.15, SE =.06). The ratings on the boy-inter nalizing video were also found to
38 be significantly higher than the girl-nonc linical, the boy-nonclinical, the boyexternalizing, and the girl-externaliz ing videos but still less than the girlFigure 1. Interaction between type of video and child gender for participantsÂ’ ratings on the Anxious-Depre ssed subscale of the CBCL. internalizing video. See Table 5 and Figure 2. Thus, participants rated the girlinternalizing video as showing more withdrawn-depressed symptoms than the boy-internalizing video but overall these two videos were rated higher on withdrawn-depressed symptoms than any of the other videos. Tukey post hoc tests showed that parti cipantsÂ’ ratings on the AggressiveBehavior subscale for the boy-externalizing video (M =1.05, SE =.07) were girl boyGender of the child in video 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Estimated Marginal Means Type of video internalizing externalizing non-clinical
39 significantly higher than the girl-nonclinical video (M =.18, SE =.08), the boyinternalizing video (M =.20, SE =.07), the girl-internalizing video (M =.29, SE =.08), and the boy-nonclinical video (M =.50, SE =.08), but not the girl-externalizing Figure 2. Interaction between type of video and child gender for participantsÂ’ ratings on the Anxious-Depre ssed subscale of the CBCL. video (M =.81, SE =.08). The ratings on the girl-ext ernalizing video were found to be significantly higher than the girl-nonc linical, boy-internalizing, and girlinternalizing video but not higher than the boy-nonclinical video. Lastly, the boynonclinical video was found to be signific antly higher than the girl-nonclinical girl boyGender of the child in video 0.00 0.25 0.50 0.75 1.00 1.25 1.50 Estimated Marginal Means Type of video internalizing externalizing non-clinical
40 video but not higher than the boy or girl internalizing videos. See Table 5 and Figure 3. Thus, the boy-externalizing and th e girl-externalizing videos were rated higher on aggressive behavio r symptoms than any of the other videos. However, while the boy-externalizing video was rated higher than the boy-nonclinical video Figure 3. Interaction between type of video and child gender for participantsÂ’ ratings on the Aggressive Behavior subscale of the CBCL. on aggressive behavior symptoms, there wa s no difference in ratings between the girl-externalizing and the boy-nonc linical videos. In order to get a global internalizing mean, exte rnalizing mean, and total mean the participantsÂ’ ratings on the internalizing subscale scores (Anxiousgirl boyGender of child in video 0.00 0.25 0.50 0.75 1.00 1.25 1.50Estimated Marginal Means Type of videointernalizing externalizing non-clinical
41 Depressed and Withdrawn-Depr essed) were combined, the participantsÂ’ ratings on the externalizing subscale scores (R ule Breaking Behavior and Aggressive Behavior) were combined, and the participa ntsÂ’ ratings on all of the subscale scores were combined. A nother 2x2x3 MANOVA was run looking at these three dependent variables. The overall MA NOVA was found to be significant (F (3,136)=158.16, p <.001). Results from this MANOVA (Table 6) showed a significant main effect for the type of video (F (6,272)=41.42, p <.001) and a significant interaction between type of video and child gender of the video (F (6,272)=3.65, p <.01). Table 6. Multivariate and univariate F va lues for Parent Gender by Type of Video by Child Gender of Video interactions fo r the Internalizing Mean, Externalizing Mean, and Total Mean of the parentsÂ’ ratings. Multivariate Univariate Variable All Internalizing Mean Externalizing Mean Total Mean Parent Gender (P) 1.25 1.07 2.99 2.74 Type of Video (T) 41.42*** 44.91*** 36.77*** 14.30*** Child Gender of Video (C) 2.14 0.76 4.01* 3.16 P by T 1.10 1.45 1.70 2.03 P by C 0.86 0.04 0.07 0.05 T by C 3.65** 2.24 3.50* 1.36 P by T by C 1.99 3.64* 2.44 3.34* *p<.05 **p<.01 ***p<.001 Follow-up univariate tests s howed that the Internaliz ing Mean, Externalizing Mean, and Total Mean were significant for the type of video ((F (2,138)=44.91,
42 p <.001), (F (2,138)=36.77, p <.001), (F (2,138)=14.30, p <.001), respectively) and the Externalizing Mean was significant fo r the interaction between type of video and child gender of the video (F (2,138)=3.50, p <.05). Tukey post hoc tests showed that the participantsÂ’ ratings on the Internalizing Mean for the internalizing videos (M =.89, SE =.04) were significantly higher than the nonclinical videos (M =.31, SE =.05) and the externalizing videos (M =.46, SE =.04). On the Externalizing Mean, participantsÂ’ rati ngs for the externalizing videos (M =.66, SE =.04) were significantly higher than the internalizing videos (M =.20, SE =.04) and the nonclinical videos (M =.28, SE =.05). On the Tota l Mean, participantsÂ’ ratings for the exte rnalizing videos (M =.59, SE =.04) were significantly higher than the internalizing videos (M =.46, SE =.04) and the nonclinical videos (M =.29, SE =.04). Additionally, the internalizing videos were rated significantly higher than the nonclincal videos on ov erall behavior problems. For the interaction between type of video and child gender of the video, Tukey post hoc tests for the Externaliz ing Mean showed that the participantsÂ’ externalizing ratings for t he boy externalizing video (M =.75, SE =.06) were significantly higher than t he boy-internalizing video (M =.16, SE =.05), the girlnonclinical video (M =.18, SE =.07), the girl-internalizing video (M =.24, SE =.06), and the boy-nonclinical video (M =.37, SE =.06) but not the girl externalizing video (M =.57, SE =.06). The girl-externalizing video was found to be significantly higher than the boy-internalizing, gi rl-nonclinical, and girl-int ernalizing video but not the boy-nonclinical video. See Figure 4. This pattern of results suggests that the externalizing videos were rated higher on externalizing symptoms than any other
43 videos and there was no difference in rati ngs of externalizing symptoms between the boy and girl-externalizing videos. Al so, while the boy-ext ernalizing video was rated higher on externalizing symptoms than the boy-nonclinical video, there was Figure 4. Interaction between type of video and child gender for participantsÂ’ ratings of the Externalizi ng Mean (items from both externalizing scales of the CBCL). no difference in ratings of externaliz ing symptoms between the boy-nonclinical video and girl-externa lizing video. girlboyGender of child in video 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 Estimated Marginal Means Type of video internalizing externalizing non-clinical
44 Note that MANOVA analyses were not Bonferroni corrected given that multivariate analyses are more conserva tive and therefore have less power than univariate tests that are Bonferroni co rrected. Therefore, by Bonferroni correcting, even more power would hav e been taken away from these analyses and the ability to detect a significant di fference when there really was one would have been lessened greatly (Nak agawa, 2004; Overall & At las, 1999). Overall, hypothesis 1 was not support ed because there were no significant differences between mothers and fathers ratings on the internalizing videos and hypothesis 2 was not supported because there was no interaction between parent gender and the type of video they saw. However, these results showed that there were interactions between the type of video the participants saw and the gender of the child in the video. ParentsÂ’ Psychological Symptoms, Inte rparental Conflict, and Own Child Ratings The third hypothesis stated that higher levels of parental psychological symptoms and higher levels of interparental conflict in mothers and fathers would predict higher ratings of ch ild behavior problems in the videos. A series of eight multiple regression analyses were conduc ted to evaluate the prediction of the Anxious/Depressed, Withdr awn/Depressed, Aggressive Behavior, and RuleBreaking Behavior subscale scores of the CBCL from levels of parental psychological symptoms (based on scores from the Brief Symptom Inventory), and levels of interparental conflict ( based on scores from the OÂ’Leary Porter Scale), for both mothers and fathers. T hus, there were four regressions for mothers and four for fathers. In addition, another series of six multiple regression
45 analyses were conducted to evaluate the prediction of the overall Internalizing Mean, Externalizing Mean, and Total Mean. Multiple regression analyses were conducted because significant beta weights would identify the unique variance of each variable to ratings of child behavior problems. Descriptive statistics for both mothers and fathers on the BSI and O PS can be found in Tables 7 and 8. Table 7. MothersÂ’ descriptive statis tics for Brief Symptom Inventory (BSI), OÂ’Leary Porter Scale (OPS), and their rati ngs of their own ch ild on the Child Behavior Checklist (CBCL). Mean SD Minimum Maximum BSI Total 0.33 0.25 0.00 1.45 OPS Total 9.26 5.20 1.00 21.00 Internalizing T 49.74 9.16 33.00 70.00 Externalizing T 46.53 7.44 33.00 61.00 Total T 47.83 7.79 31.00 64.00 Table 8. FathersÂ’ descriptive statis tics for Brief Symptom Inventory (BSI), OÂ’Leary Porter Scale (OPS), and their rati ngs of their own ch ild on the Child Behavior Checklsist (CBCL). Mean SD Minimum Maximum BSI Total 0.24 0.23 0.00 1.04 OPS Total 9.03 5.07 2.00 27.00 Internalizing T 47.89 10.12 31.00 69.00 Externalizing T 46.65 8.38 33.00 69.00 Total T 46.79 9.28 26.00 71.00 Results from the multiple regression analyses can be found in Tables 9 and 10, for the four subscales and the three summa ry means, respectively. Overall, the third hypothesis was not supported. Pa rticipantsÂ’ level of psychological
46 Table 9. Multiple regression analyses of mothersÂ’ and fathersÂ’ ratings on the Brief Symptom Inventory (BSI) and OÂ’Leary Porter Scale (OPS) predicting ratings of the videos across four behavior subscales. Mothers Fathers Dependent Variable B Std. Error p B Std. Error p Anxious-Depressed BSI OPS 0.02 0.01 0.20 0.01 0.94 0.35 -0.11 -0.01 0.21 0.01 0.59 0.34 Withdrawn-Depressed BSI OPS -0.19 -0.00 0.28 0.01 0.51 0.83 -0.23 -0.03 0.31 0.01 0.47 0.06 Rule Breaking BSI OPS 0.12 -0.00 0.12 0.01 0.35 0.78 -0.02 0.00 0.15 0.01 0.91 0.93 Aggressive Behavior BSI OPS 0.11 0.00 0.23 0.01 0.63 0.92 -0.10 0.01 0.29 0.01 0.75 0.56 Table 10. Multiple regression analyses of mothersÂ’ and fathersÂ’ ratings on the Brief Symptom Inventory (BSI) and OÂ’Leary Porter Scale (OPS) predicting ratings of the videos across Internalizing Mean, Externalizing Mean, and Total Mean. Mothers Fathers Dependent Variable B Std. Error p B Std. Error p Internalizing Mean BSI OPS -0.06 0.00 0.20 0.01 0.77 0.65 -0.16 -0.02 0.20 0.01 0.44 0.09 Externalizing Mean BSI OPS 0.11 0.00 0.17 0.01 0.50 0.97 -0.06 0.00 0.21 0.01 0.79 0.66 Total Mean BSI OPS 0.05 0.00 0.13 0.01 0.72 0.82 -0.09 -0.00 0.17 0.01 0.57 0.67
47 symptomatology and level of interparental conflict in their home were not related to their ratings of the child Â’s behavior in the video. The fourth hypothesis stat ed that higher ratings of a parentÂ’s own childÂ’s behavior problems would be related to highe r ratings of behavior problems of the child in the video. Pearson correlati ons were conducted to look at the relationship between participantsÂ’ ratings of their own child on the CBCL (Internalizing T, Externalizing T, and Total T) and the mean behavior ratings of the child in the video (Internalizing Mean, Externalizing Mean, and Total Mean). When mothers were looked at separately, there was a significant correlation between mothersÂ’ Total Mean ratings of the child in the video and the Internalizing T score of their own child on the CBCL (r (77)=.34, p <.01). There also was a significant correlation between mothersÂ’ Total Mean ratings of the child in the video and the Total T score of their own child on the CBCL (r (77)=.23, p <.05). Lastly, there was a significant correlation betw een mothersÂ’ Externalizing Mean ratings of the child in the video and the Total T score of their own child on the CBCL (r (77)=.25, p <.05). In contrast, there we re no significant correlations between fathersÂ’ ratings of the child in the video and their ratings of their own child. FischerÂ’s z tests were perform ed to see if there were significant differences between the mothersÂ’ correlation s of their ratings of the child in the video and their own child and the fathersÂ’ co rrelations. These comparisons were non-significant (all pÂ’s > .05). Theref ore, although mothers did show some association between their ratings, they did not differ significantly from fathers. Pearson correlations were performed for t he entire sample to see if there were
48 significant correlations between both mothersÂ’ and fathersÂ’ ratings of the child in the video and their own child. As can be seen in Table 11, none of these correlations were significant. In short, the proposed hypotheses we re not supported. However, there were significant differences in ratings of the child in the video based on the type of video the participant saw and the gender of the child in the video. Table 11. Pearson correlations of parentsÂ’ ratings of the child in the video on items from four subscales of the CBCL (Internalizing Mean, Externalizing Mean, and Total Mean) and the parentsÂ’ ratings of their child closest in age to 6 on the CBCL (Internalizing T-score, Externa lizing T-score, and Total T-score). 1 2 3 4 5 6 1. Internalizing Mean .14 .63** .10 .02 -.00 2. Externalizing Mean .86** .10 .01 .11 3. Total Mean .14 .02 .08 4. Internalizing T .35** .76** 5. Externalizing T .74** 6. Total T **p<.01 Note: Variables 1-3 are for parentsÂ’ rati ngs of the video and variables 4-6 are for parentsÂ’ ratings of their own child. Bold signifies the results of the hypotheses tested.
49 Discussion Lack of Mother-Father Differences in Ratings Overall, the hypotheses regarding discrepancies between mothersÂ’ and fathersÂ’ ratings were not supported. Thus, when mothers and fathers have equivalent rates of contact with a child (in this case, no contact) they do not differ in their ratings of child behavior. Be cause of the experimental design of the study, using a child in a video that the participants did not know and had no contact with previously, the results l end support to the idea that differential contact of mothers and fathers with their childr en could be related to disagreement about child behavior problems in their own children. At the same time that similar contact or lack thereof could be related to stronger agreement. Duhig et al. (2000) repo rted that mothers tended to report more behavioral problems than fathers due to the greater amount of cont act. By being more involved in child rearing and having more awareness and insight into their childÂ’s behaviors, mothers appeared to overr eport child behavior problems in comparison to fathers (Christensen et al., 1992). Also, Schaughency and Lachey (1985) and Seiffge-Krenke and Kollmar (1998) st ated that fathers lack accuracy in their ratings due to the lower amount of time that they interact with their children. Therefore, the inaccuracy of paternal ratings would lead to discrepant ratings between parents. In contrast to these studies, in the present study,
50 mothers and fathers had the same amount of Â“contactÂ” or viewing time of the child in the video, and therefore, cont act was not a differential factor. When rating their own children, Rowe and Kandel (1997) stated that parents may have differing access to samples of their childrenÂ’s behavior, a term Achenbach et al. (1987) term ed Â“situational specificity. Â” Also, fathers may see more appropriate behavior given that child ren obey their fathers more frequently and therefore, fathers might rate their child different ly than mothers (Campbell, 1991). In the present study, not only did participants vi ew the child in similar situations but participants did not have to rely on retrospective ratings but rather were able to rate the childÂ’s behaviors immediately after viewing the behaviors. Therefore, there was less ti me to have factors, such as psychological symptoms and stress, distort ratings. Kroes, Vee rman, and De Bruyn (2005) reported that being familiar with a target that you are rating influences the over-reporting of behavioral problems, perhaps because one has greater access to many instances of behavior rather than one distinct episode or event. Therefore, the greater amount of access that mothers t end to have with their childrenÂ’s behavior may inflate their ratings of their own childr en. However, in the present study both mothers and fathers had the same amount of access to the behavior of the child in the video and this fact could have contributed to the lack of differences between mothersÂ’ and fathersÂ’ ratings. Although many studies have found differ ences in mothersÂ’ and fathersÂ’ ratings of child behavior problems, tw o major meta-analyses have shown these differences to be small and insignificant (Achenbach, et al., 1987; Duhig et al.,
51 2000) and many other studies have found moderate to high parental correspondence (Hay et al., 1999; Jens en et al., 1988; Rowe & Kandel, 1997; Seiffge-Krenke, 1998; Webster-Stratton, 1988). Achenbach et al. (1987) found that parents did not differ across ratings for externalizing and internalizing behavior problems of their child. In addition, they stated that there was a higher correspondence for six to eleven year ol d children than there was for adolescent children. In the current study, the ch ild in the video was eight years old. However, Duhig et al. (2000) found that when rating internalizing and externalizing behavior pr oblems, greater correspondence was found for adolescents rather than for younger children. Therefore, evidence for the effect of child age is inconclusive. Duhig et al (2000) also found that correlations were significantly higher for informants in similar roles who recorded behavior simultaneously. Likewise, participants in t he present study rated the child in the video by viewing the child in a similar context and in a role as an outside observer. Epkins (1996) stated that scales that are equivalent or parallel across informants also lead to bette r agreement. That was also the case in the present study in which participants rated the ch ild in the video on the same 56 behavioral items from the CBCL. Duhi g et al. (2000) f ound that overall mothers and fathers displayed very small and insignificant diffe rences in their ra tings of childrenÂ’s behavior problems. In looking at moder ators that may infl uence ratings, higher socioeconomic status of the parents was found to be associated with greater correspondence between mothers and fathers. Similarly, in the present study
52 where socioeconomic status of the parents was relative ly high, there were no differences between mothers and fathers in ratings of childrenÂ’s behavior. Several studies have also found a lack of difference between mothersÂ’ and fathersÂ’ ratings of child behavior. An early study by Thompson and McAdoo (1973) found no differences between mother sÂ’ and fathersÂ’ ratings of their children across seven subscales of the Mi ssouri ChildrenÂ’s Behavior Checklist. Guerney, Shapiro, and Stover (1968) found low to moderate correlations between parentsÂ’ ratings of their mal adjusted children on a problem list of behaviors and an interpersonal list of behaviors. Although correlations ranged from .32 to .74, all were signific ant except one and the magnitude of the correlations showed a strong degree of agr eement. Webster-Stratton (1988) also found no differences between mother sÂ’ and fathersÂ’ ratings on the CBCL, the same measure used in the present study. Lastl y, Rowe and Kandel (1997) found that parental ratings of children contained a subs tantial trait component in which a large amount of variance in rati ngs was shared amongst parents, rather than an individual view in which varianc e was unique to only one rating source. This example also supports the idea that parents do correspond on ratings of children. Lack of Influence of Psychological Symptoms and Interparental Conflict The present study did not find that hi gh levels of parental psychological symptoms and interparental c onflict predicted higher rati ngs of the child in the video. Despite research on the distorti on hypothesis and other studies that have shown parentsÂ’, most often mothersÂ’, ratings were influenced by their own
53 psychopathology (Chi & Hinshaw, 2002; P hares et al., 1989; Richters, 1992), other studies have found a lack of influenc e. In Kroes et al.Â’s (2005) study, parents watched videotapes of children, so me of whom they were familiar with and some whom they were not. When one was acquainted with the child, they reported more problems. However, mothersÂ’ psychological symptoms and stress did not play a factor as expected gi ven that the video created an emotional distance and reduction of stre ss. Therefore, the lack of impact of their own traits appeared to reduce mothersÂ’ ratings. Sim ilarly, in this study, psychological symptoms and interparental conflict ma y not have played a role given the structure and experiment al design of the study. Some studies have found the effect of psychological symptoms and interparental conflict to be a factor for ma ternal ratings but not paternal ratings. Webster-Stratton (1988) found that mot hersÂ’ ratings were influenced by low marital satisfaction and negativity but not fa thersÂ’ ratings. It was suggested that mothers may have felt more guilt and stre ss related to their own parenting role, especially given the high leve l of conduct problems in thei r children in the study. In contrast, fathers may not have felt as guilty and may have dealt with their stress differently from mot hers. However, because participants in the present study rated a child that was not thei r own, stress and guilt over their own parenting role should not have been a fa ctor. Thus, there would be little influence of psychological symptoms and conflict on participantsÂ’ ratings. Seiffge-Krenke and Kollmar (1998) also found similar results in which mothers who experienced stress inflicted by marital problems perceived greater
54 symptoms in their children than fathers w ho were relatively unaffected by their own personal adjustment. In addition, Ha y et al. (1999) f ound that mothersÂ’ reports were more influenced by their own mental state and view of their marriage while fathersÂ’ reports were bas ed on the childÂ’s cognitive ability. FathersÂ’ reports tended to correspond more accurately with teachersÂ’ reports. Therefore, mothers seem to be more influenced by their own psychological symptoms and marital conflict and this pattern of results may explain why fathersÂ’ psychological symptoms and interparent al conflict did not predict higher behavioral ratings. For mothers in t he present study, however, a lack of significant findings as m entioned before may have been due to the distant nature of the child in the video who did not evoke the same stress and guilt about their parenting role had the child been their own. While mothersÂ’ ratings in the present study were not influenced by their own psychological symptoms and interpar ental conflict, there was an association between their ratings of the child in the vi deo and their ratings of their own child. Connell and Goodman (2002) pointed out that within families there exist bidirectional influences in that a child Â’s psychopathology may lead to a parentÂ’s psychopathology and stress at the same time that a parentÂ’s psychopathology may lead a child to exhibit more negative functioning. If oneÂ’s child has a high level of dysfunction, then there is added st ress to the family regardless of where it originated. Given that mothers are more influenced by psychological stress and conflict (Webster-Stratton, 1988), then t he mothers who saw the child in the video act in a distressing manner may have been influenced to rate their own
55 child higher. Or, rather than pinpointing mothersÂ’ distress to their own psychological symptoms, it could be that their childÂ’s own psychological symptoms influenced their ratings of other children, such that they saw another child in a more negative light. Because the participants saw the video before they rated their own child, the former ex planation is more probable. Murray and Sacco (1998) found that when a mother held a negative conception of a child, she was more likely to make negative affective reactions to child behavior. Therefore, future research should look into how viewing a video of a child portraying negative behaviors, influences how one rates their own child. Child Gender Differences Several significant interactions we re found for the type of video and the gender of the child in the video. Therefor e, there existed a relationship between how participants viewed boys versus girls and the types of behaviors the children displayed. The first interaction betw een type of video and child gender in the video was found significant for participant sÂ’ ratings on the Anxious-Depressed subscale, such that the boy-internalizi ng video was rated higher than any of the other videos but not signific antly higher than the girl -externalizing video. The similarity in ratings on the Anxious-D epressed scale for the boy-internalizing video and girl-externalizing video speaks to how parents rate boys versus girls on these items. The Anxious-Depressed scale has more action oriented and aggressive items than the Wi thdrawn-Depressed scale of the CBCL. Such items include, Â“fears he/she might think or do something bad,Â” Â“nervous, highstrung, or tense,Â” Â“talks about killing self,Â” Â“cri es a lot,Â” and Â“fears certain animals,
56 situations, or places.Â” In contrast Withdrawn-Depressed items include more passive and internal items like, Â“too sh y or timid,Â” Â“withdrawn, doesnÂ’t get involved with others,Â” Â“unhappy, sad, or depressed,Â” and Â“there is very little he/she enjoys.Â” Research has shown that the it ems on the ChildrenÂ’s Depression Inventory (CDI), another m easure of internalizing behavior, have been criticized because they seemed to be tapping into f eatures other than depression (Liss, Phares, & Liljequist, 2001). More specific ally, certain items may be measuring features of aggression and externalizing behaviors in addition to depressive symptoms. In addition, the CDI was unable to distinguish between children diagnosed with internalizi ng disorders versus children with externalizing disorders. Therefore, t he Anxious-Depressed subsca le may also be measuring features of externalizing disorders and may lead participants to rate boys higher than girls on such items. More classic externalizing symptoms, like in the case of AttentionDeficit/Hyperactivity Disorder (ADHD) feat ures such as impulsivity, aggression, and inattentiveness, tend to be rated higher in boys whereas girls tend to be rated higher on more indirect relati onal aggression (Jackson & King, 2004). Therefore, participants ma y have rated the girl-ext ernalizing video higher on Anxious-Depressed features given the more aggressive nature of the items and the fact that girls do not normally displa y classic externaliz ing symptoms to the extent that boys do. Gender-role research has shown that girls are socialized to express internal emotions such as s adness and empathy and to inhibit external
57 emotions like anger or aggressiveness. On the contrary, bo ys are socialized to repress emotions that make them look vulnerable and express those that make them appear more powerful and in cont rol (Timmers et al., 1998). Due to socialization factors, part icipants may have viewed the boy-internalizing video as possessing more features of the Anxious -Depressed scale and more typical to boysÂ’ display of internalizing symptom s (more outwardly and aggressive) rather than girlsÂ’ internalizing symptoms that may be more inward and passive in nature. Likewise, in the interaction of child gender and type of video for the Withdrawn-Depressed subscale, the girl-internalizing video was rated significantly higher than the boy-internaliz ing video. Therefore, even though the boy and girl showed the same behaviors in the video, participants viewed their behavior differently. In Morrongiello and Da wberÂ’s (2000) study in which mothers viewed a videotape of a boy and girl engag ing in similar risk-taking behavior, parents encouraged boysÂ’ risk taking behavio r and cautioned girlsÂ’ risk taking behavior. Therefore, girls may be more likely to internalize what can happen in risk taking behaviors more than boys and may be more aware of their vulnerability. These internalizing characteristics are more accepted in girls than boys and could contribute to higher ratings of girlsÂ’ internalizing symptoms than boysÂ’ internalizing symptom s. Seiffge-Krenke and Kollm ar (1998) stated that males are less inclined to discuss private worries and problems and to reveal emotional stress and therefore, parents are less aware of how sons are coping emotionally. If parents ar e not aware then these behaviors may be perceived as
58 less common in boys and therefore, less acc epted. Participants in the present study may have seen the girl as po ssessing more stereotypical female internalizing characteristics and thus, may have rated her higher than the boy displaying the same behavior. However, Jensen, Traylor, Xenakis, and Davis (1988) stated that when girls display less characteristic and stereotypical behaviors, such as externalizing sympto ms, they will be rated higher than boys because the behavior is less tolerated. Al so, teachersÂ’ expectancies were found to vary by gender. Specifically, gi rls did not need to portray as many externalizing behaviors to be diagnosed with ADHD as boys did because the behavior is less frequent in girls and is not as socially acceptable (Jackson & King, 2004). Therefore, evidence is inconc lusive and needs further research into gender-stereotypes and ratings of childrenÂ’s behavior. Block (1973) stated that socializat ion is a primary reason why boys and girls act the way they do. Boys are taught to control their feelings and girls are taught to express their emot ions and concerns for others. Boys are taught to be assertive and independent while girls are taught to control this assertion. Also, parenting styles can be instrumental in developing these stereotypical behaviors (Maccoby, 1998). Parents tend to handl e their daughters more gently and ask them about their feelings. On the other hand, parents are more tolerant of fighting with their sons and more likely to use physical punishment. Thus, a boyÂ’s play is more likely to be rougher and centered around physical and outward behaviors while girlsÂ’ play is more likely to be centered around emotions and internalizing behaviors. Recently, Diamantopoulou, Henr icsson, and Rydell
59 (2005) found that even peers of children adhere to what they call the Â“gender appropriateness hypothesisÂ” in which they tole rated higher levels of externalizing behaviors in boys rather than girls. Overa ll, females do not express all emotions more than males but some emotions ar e more likely in females, such as happiness, sadness, fear, guilt, and shame. In contrast, males are more likely to express anger, pride, and cont empt (Brody & Hall, 1993) Hence, socialization differences and the adherence to stereoty pes in which emotional and inward reflecting behaviors are reinforced and socialized in girls and outward and physical behaviors are reinforced and em otions are repressed in boys may reflect why participants rated the girl higher than the boy on internalizing behaviors and why boys tend to be rated hi gher on externalizing behaviors than girls. In the final two interactions, even though the boy-externalizing video was rated highest, there was no significant difference between the participantsÂ’ ratings on Aggressive-Behavior subscale and the overall Externalizing Mean for the boy and girl-externalizing video. Ho wever, while the boy-externalizing video was rated higher than t he boy-nonclinical video, there was no significant difference between the girl-externalizing video and the boy-nonclinical video. Socialization and gender role differ ences may account for why the boyexternalizing video was rated higher t han all of the other videos on these externalizing scales. M aniadaki, Sonuga-Barke, and Ka kouros (2005) recently found that boys with ADHD show more ex ternalizing and disruptive behaviors than girls and that parents attribute t hese behaviors as more intentional and
60 therefore impose stricter responses. Ther efore, parents would be more likely to see the boyÂ’s behavior as more severe and rate it accordingly. Similarly, several studies have found that boys tend to be rated higher and exhibit more overall symptoms than girls (Duhig et al., 2000; Jackson & King, 2004; Seiffge-Krenke & Kollmar, 1998). As Christensen et al. (1992) pointed out, parents may pay closer to attention to boys overall and discu ss boysÂ’ behavior at greater length than girlsÂ’ behavior. Therefore, regardl ess of the type of behavior the boy is displaying, it will be rated higher than girls. Also, because externalizing behavior is more accepted and expected in boys, paren ts may be more likely to think they remember seeing those behaviors when reporting because they are more prototypical. This phenomenon is related to social schemas, as first proposed by Bartlett (1932), about how males are suppos ed to act and therefore affects the information that is recalled. This res earch would explain the lack of difference between the girl-externalizing and boy-nonc linical videos and would explain why boys were rated higher on male stereotyp ical externalizing symptoms regardless of the behaviors they portrayed. In short, socialization and gender-ro le differences account for why participants viewed the videos that portray ed a girl and boy engaging in similar behaviors so differently. Implications for these findings suggest that even if parents agree on childrenÂ’s behavi ors there still may be some biases in how they view boysÂ’ versus girlsÂ’ behavior. Pr ofessionals should be aware of these gender stereotypes and take them into acc ount when parents, teachers, and other professionals are rating children. Is a girl being rated higher on internalizing
61 symptoms simply because that is more appropriate female behavior or does the girl truly present distressing behaviors in need of treatment? Likewise, is a boy being rated higher on externalizing symptoms simply because he is displaying stereotypical acting out behavior or does this boy show dysfunctional behavior that is in need of school and psychosocia l interventions? These questions and others should be looked into further. Limitations and Future Research There were several limitations to this study. First, the sample was primarily comprised of parti cipants with medium to high so cioeconomic status. In addition, the sample was primarily Caucas ian. Future res earch would benefit from looking at parental ratings of ch ild behavior problems with a lower SES and more diverse ethnic and racial populati on. Second, the participants were primarily married. However, the res earchers did not include a choice on the demographic form to indicate whether t hey were in their first marriage or had been remarried. Therefore, it is not known if the participants had ever been divorced or separated in the past. Additionally, the lack of single or di vorced parents may have been related to the relatively low ratings of interpar ental conflict, psycholog ical symptoms, and their own child behavior problems. Epst ein et al. (2004) r eported that using a community sample as in the present study, the mean maternal ratings of interparental conflict on the OPS wa s 12.24 and mean paternal ratings was 12.12. In the present study, mean mater nal and paternal ratings were quite a bit lower (9.26 and 9.03, respectively). T herefore, lower ratings may have been due
62 to the majority of the sample being married. In addition, all of the parental ratings of their own child were well below t he borderline-clinical level of behavior problems on the CBCL (T-score=60). Mate rnal mean of the total T-score was 47.83 and paternal mean of the total T-score was 46.79. Lastly, the maternal mean on the BSI was .33 and the paternal mean was .24. These results demonstrate minimal psychological symptoms across the sample and are comparable to the Adult Nonpatients from the BSI Normative sample (Females, M=.35; Males, M=.25; Derogatis, 1993). The low ratings of psychological symptoms, interparental conflict, and child behavior problems may have been associated with why these factors did not predict higher behavior ratings of the child in the video. Future research woul d benefit from looking at parental ratings of a child in a video, or one they do not k now, in a clinical rather than community sample. The influence of psychologica l symptoms and inter parental conflict on ratings of childrenÂ’s behavior may be more pronounced in a clinical sample due to the high prevalence of these problems. Other limitations in the cu rrent study include the length of the video. The video was only eight minutes in length and therefore did not cover a wide range of symptoms and perhaps did not allow enough time to gauge the full range of the childÂ’s functioning. Therefore, par ental ratings of the child may have been lower than if they viewed the child in the video for a longer period of time. Future research should look at parentsÂ’ ratings of a child who is videotaped for a longer period of time or observed in a natura listic setting. By changing the surroundings and the length of time, parents would be ab le to get a better range of behaviors
63 and be able to make more accurate ratings Another limitation was that both children in the videos were White and around the age of 8 years old. Therefore, some parents may not have identified with the child in the video due to having children of differing races and ages. This may have been associated with less accurate ratings. Future research w ould benefit from in cluding children with multiple races and ages to see if parental ratings are different depending upon race and age. Lastly, only parents were included in this sample. Therefore, future research would benefit from includi ng non-parents to see if they differ from parents in how they view childrenÂ’s behav ior. Perhaps, by not having contact with children of their own non-parents may take a more objective and accurate view of other childrenÂ’s behavior. Conclusions In the present study, mot hers and fathers did not differ on their ratings of a child with whom they had no contact and di d not know. Theref ore, knowing oneÂ’s child and having a range of behaviors and experiences to reflect upon may influence parentsÂ’ ratings and lead to mo re disagreement about their childÂ’s overall behavioral problems. In addition mothersÂ’ and fathersÂ’ psychological symptoms and interparental confli ct were not related to thei r ratings of the child in the video. Perhaps, due to t he distant nature of the video and rating a child they did not know, their own problems were less likely to be related to how they rated a child who was not their own. Ther e was an association between mothersÂ’ ratings of the child in the video and thei r own child. Therefore, instead of being related to their own psychological symptom s, mothersÂ’ ratings were more related
64 to their own childÂ’s symptoms. Future research would b enefit from looking further at how oneÂ’s own child is relat ed to how parents see other children. Lastly, parents did view the children in the video very differently dependent upon the childÂ’s gender and the type of behavior t he child portrayed. Future studies should examine the influence that stereot ypical and non-stereotypical behaviors have on parental ratings. Overall, it is important for clinicians, parents, and other professionals to be aware of how factors, like the amount of c ontact, the type of behavior, and a childÂ’s gender, affect ratings of childrenÂ’s behavior. As always, gathering information from multiple in formants will provide the best assessment of a childÂ’s functioning.
65 List of References Achenbach, T.M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: University of Ve rmont, Department of Psychiatry. Achenbach, T.M. (1992). Manual for the Child Behavior Checklist/2-3 and 1992 Profile. Burlington, VT: University of Ve rmont, Department of Psychiatry. Achenbach, T.M., McConaughy, S.H., & Howe ll, C.T. (1987). Child/adolescent behavioral and emotional problems: Im plications of cross-informant correlations for situational specificity. Psychological Bulletin, 101, 213222. Achenbach, T.M. & Resco rla, L.A. (2001). Manual for the ASEBA School-Age Forms & Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. Bartlett, F.C. (1932). Remembering: An Experimental and Social Study. Cambridge: Cambridge University Press. Bingham, C.R., Loukas, A., Fitzgerald, H. E., & Zucker, R.A. (2003). Parental ratings of sonÂ’s behavior problems in high-risk families: Convergent validity, internal structur e, and interparent agreement. Journal of Personality Assessment, 80, 237-251. Block, J.H. (1973). Concepti ons of sex role: Some cro ss-cultural and longitudinal perspectives. American Psycholoogist, 2, 512-526. Brody, L.R. & Hall, J.A. (1993). Gender and emotion. In M. Lewis & J.M. Haviland (Eds.), Handbook of emotions (pp.447-460). New York: Guilford Press. Campbell, S.B. (1991). Behavior problems in presc hool children: Clinical and developmental issues. New York: Guilford Press. Chi, T.C. & Hinshaw, S.P. (2002). Mot her-child relationships of children with ADHD: The role of maternal depre ssive symptoms and depression-related distortions. Journal of Abnormal Child Psychology, 30, 387-400. Christensen, A., Margolin, G., & Sullaway M. (1992). Interparental agreement on child behavior problems. Psychological Assessment, 4, 4, 419-425.
66 Christensen, A., Phillips, S., Glascow, R. E., & Johnson, S.M. (1983). Parental characteristics and interactional dysf unction in families with child behavior problems. A preliminar y investigation. Journal of Abnormal Child Psychology, 153-166. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159. Connell, A. & Goodman, S. (2002). The association between psychopathology in fathers versus mothers and childrenÂ’ s internalizing and externalizing behavior problems: A meta-analysis. Psychological Bulletin, 128, 746-773. Costa, P.T. & McCr ae, R.R. (1992). Revised NEO personality inventory (NEO PI-R) and NEO five-factor inventory (NEO-FFI). Odessa, FL: Psychological Assessment Resources. Derogatis, L. (1993). Brief Symptom Inventory (BSI) Administration, Scoring, and Procedures Manual (3rd ed.). Minneapolis, MN: National Computer Systems, Inc. Diamantopoulou, S., Henricsson, L., & Rydell, A. (2005). ADHD symptoms and peer relations of children in a community sample. International Journal of Behavioral Development, 29, 388-398. Dodge, K.A. (1986). A social information processing model of social competence in children. In M. Perlmutter (Ed.), Minnesota symposium on child psychology (Vol. 18, pp. 77-125), Hillsdale, NJ: Erlbaum. Duhig, A., Renk, K., Epstein, M., & P hares, V. (2000). In terparental agreement on internalizing, externalizing, and total behavior problems: A metaanalysis. Clinical Psychology: Science and Practice, 7, 435-453. Epkins, C.C. (1996). Parent ratings of childrenÂ’s depression, anxiety, and aggression: A cross-sample analysis of agreement and differences with teacher ratings. Journal of Clinical Psychology, 52, 599-608. Epstein, M.K., Renk, K., Duhig, A.M. Bosco, G.L., & Phares, V. (2004). Interparental conflict and adoles cent adjustment: Convergent and discriminant validity. Educational and Psychological Measurement, 64, 475-495. Frosch, C. & Mangelsdorf, S. (2001). Marital behavior, parent ing behavior, and multiple reports of preschoolersÂ’ behavior problems: Mediation or moderation? Developmental Psychology, 37, 502-519.
67 Guerney, B.G., Shapiro, E.B., & Stover, L. (1968). Parental perceptions of maladjusted children: Agreement bet ween parents and relation to motherchild interaction. The Journal of Genetic Psychology, 113, 215-225. Hay, D.F., Pawlby, S., Sharp, D., Schmucker, G., Mills, A ., Allen, H, & Kumar, R. (1999). ParentsÂ’ judgments about young childrenÂ’s problems: Why mothers and fathers might disagree yet still predict later outcomes. Journal of Child Psychology and Psychiatry, 40, 1249-1258. Hollingshead, A.B. (1975). Four Factor Index of Social Status New Haven, CT: Department of Sociology, Yale University. Jackson, D.A. & King, A.R. (2004). Gender differences in the effects of oppositional behavior on teacher ra tings of ADHD symptoms. Journal of Abnormal Child Psychology, 32, 215-224. Jensen, P.S., Traylor, J., Xenakis, S.N., & Davis, H. (1988). Child psychopathology rating scales and interrater agreement: I. parentsÂ’ gender and psychiatric symptoms. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 442-450. Kane, P., & Garber, J. (2004). The relations among depression in fathers, childrenÂ’s psychopathology, and father-c hild conflict: A meta-analysis. Clinical Psychology Review 24 339-360. Kroes, G., Veerman, J.W., & DeBruyn, E. (2003). Bias in parental reports? Maternal psychopathology and the repor ting of problem behavior in clinicreferred children. European Journal of Psychological Assessment, 19, 195-203. Kroes, G., Veerman, J.W., & DeBruyn, E. (2005). The impact of the big five personality traits on reports of ch ild behavior problems by different informants. Journal of Abnormal Child Psychology, 33, 231-240. Kurdek, L.A. (2003). Corre lated of parentsÂ’ perceptio ns of behavioral problems in their young children. Applied Developmental Psychology, 24, 457-473. Liss, H., Phares, V., & Liljequist, L. ( 2001). Symptom endorsement differences on the childrenÂ’s depression inventory with children and adolescents on an inpatient unit. Journal of Personality Assessment, 76, 396-411. Maccoby, E.E. (1998). The two sexes: Growing up apart, coming together. Cambridge, MA: Harvard University Press.
68 Macintyre, S. (1993). G ender differences in the perception of common cold symptoms. Social Science and Medicine, 36 15-20. Maniadaki, K., Sonuga-Burke, E., & Kak ouros, E. (2005). ParentsÂ’ causal attributions about attention deficit/hyperac tivity disorder: The effect of child and parent sex. Child: Care, Health, and Development, 31, 331-340. McConaughy, S.H. & Achenb ach, T.M. (2001). Manual for the Semistructured Clinical Interview for Ch ildren and Adolescents (2nd ed.). Burlington, VT: University of Vermont, Research Cent er for Children, Youth, & Families. Mechanic, D. (1976). Sex, illness, illness behavior, and the use of health services. Journal of Human Stress, 2, 29-40. Moretti, M.M., Fine, S., Haley, G., & Marriage, K. (1985) Childhood and adolescent depression: Child-report versus parent-report information. Journal of the American Academy of Child and Adolescent Psychiatry, 24, 298-302. Morrongiello, B.A. & Dawber, T. (2000) MothersÂ’ responses to sons and daughters engaging in injury-risk behaviors on a playground: Implications for sex differences in injury rates. Journal of Experimental Child Psychology, 76, 89-103. Murray, D. & Sacco, W.P. ( 1998). Effect of child-releva nt cognitions on motherÂ’s mood: The moderating effect of child-trait conceptions. Cognitive Therapy and Research, 22, 47-61. Najman, J.M., Williams, G.M., Nikles, J., Spence, S., Bor, W., OÂ’Callaghan, M., LeBroque, R., & Anderson, M. (2000). MothersÂ’ mental illness and child behavior problems: Cause-effect asso ciation or observational bias? Journal of the American Academy of Child and Adolescent Psychiatry, 39, 592-602. Nakagawa, S. (2004). A farewell to Bonf erroni: the problems of low statistical power and publication bias. Behavioral Ecology, 15, 1044-1045. Overall, J.E. & Atlas, R. S. (1999). Power of univariate and multivariate analyses of repeated measurements in c ontrolled clinical trials. Journal of Clinical Psychology, 55, 465-485. Phares, V. (1992). WhereÂ’s poppa? The relati ve lack of attention to the role of fathers in child and ad olescent psychopathology. American Psychologist, 47, 656-664.
69 Phares, V., & Compas, B.E. (1990). AdolescentsÂ’ subjec tive distress over their emotional/behavioral problems. Journal of Consulting and Clinical Psychology, 58, 596-603. Phares, V., Compas, B.E., & Howell, D.C. (1989). Perspectives on child behavior problems: Comparisons of ch ildrenÂ’s self-reports with parent and teacher reports. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 68-71. Phares, V., Lopez, E., Fields, S., Kambouk os, D., & Duhig, A.M. (2005). Are fathers involved in pediatric p sychology research and treatment? Journal of Pediatric Psychology. Porter, B. & OÂ’Leary, K.D. (1980). Marital discord and child behavior problems. Journal of Abnormal Child Psychology, 8, 287-295. Richters, J.E. (1992). Depr essed mothers as informant s about their children. A critical review of the evidence for distortion. Psychological Bulletin, 112, 485-499. Rowe, D.C. & Kandel, D. (1997) In the eye of the behol der? Parental ratings of internalizing and externalizing symptoms. Journal of Abnormal Child Psychology, 25, 265-275. Schaughency, E.A. & Lahey B.B. (1985). MothersÂ’ and fathersÂ’ perceptions of child behaviour, parental depressi on, and marital satisfaction. Journal of Consulting and Clinical Psychology, 53, 718-723. Seiffge-Krenke, I. & Kollmar, F. ( 1998). Discrepancies between mothersÂ’ and fathersÂ’ perceptions of sonsÂ’ and daughtersÂ’ problem behaviour: A longitudinal analysis of parent-adolesc ent agreement on internalizing and externalizing problem behaviour. Journal of Child Psychology and Psychiatry, 39, 687-697. Stanger, C. & Lewis, M. (1993). Agreement among parents, teachers, and children on internalizing and externalizing problems. Journal of Clinical Child Psychology, 22, 107-115. Thompson, R.J. & McAdoo, W. G. (1973). A comparison of mothersÂ’ and fathersÂ’ behavior checklist ratings of outpatient boys and girls. Journal of Community Psychology, 1, 387-389. Timmers, M., Fischer, A.H., & Manstead, A.S.R. (1998). Gender differences in motives for regulating emotions. Personality and Social Psychology Bulletin, 24, 974-985.
70 Webster-Stratton, C. (1988). MothersÂ’ and fathersÂ’ perceptions of child deviance: Roles of parent and child beh aviors and parent adjustment. Journal of Consulting and Clinical Psychology, 56, 909-915. Webster-Stratton, C. (2002). The rela tionship of marital support, conflict, and divorce to parent perceptions, behaviors, and childhood conduct problems. Journal of Marriage and the Family, 51, 417-430.
72 Appendix A Sample questions from the Â“Semistruc tured Clinical Interview for Children and Adolescents (SCICA) Ages 6-18 Protocol Form.Â” These items were used to interview the child actor for the videotape stimulus. Activities: 1. What do you like to do in your spar e time, like when youÂ’re not at school? Do you participate in any sports/hobbies/clubs? School: 2. Do you ever get in trouble in schoo l? Do you ever worry about school? 3. If you could change one thing abo ut school, what would it be? Friends: 4. What do you do with your friends? Do they come to your house? Do you go to their house? How often? Family Relations: 5. Who are the people in your fa mily? Who lives in your home? 6. Who makes the rules in your home? What happens when kids break the rules? Do you think the rules are fair or unfair? Self Perception, Feelings: 7. Tell me a little more about yourse lf. What makes you happy? What makes you sad? What do you do when youÂ’re sad? What makes you mad? What do you do when youÂ’re mad? What makes you scared? What do you do when youÂ’re scared?
73 Appendix B Selected Child Behavior Checklist Items (broken down by narrowband scales). Participants rated the child in the videotape on these items. 0=Not true 1=Somewhat True 2=Very True Externalizing Subscales: Rule Breaking Behavior 1. Drinks alcohol witho ut parentsÂ’ approval (describe): _____________________ 2. DoesnÂ’t seem to feel guilty after misbehaving 3. Breaks rules at home, school, or elsewhere 4. Hangs around with ot hers who get in trouble 5. Lying or cheating 6. Prefers being with older kids 7. Runs away from home 8. Sets fires 9. Sexual problems (describe): ____________________ 10. Steals at home 11. Steals outside the home 12. Swearing or obscene language 13. Thinks about sex too much 14. Smokes, chews, or sniffs tobacco 15. Truancy, skips school 16. Uses drugs for nonmedical purposes ( donÂ’t include alcohol or tobacco) (describe): _____________________ 17. Vandalism Aggressive Behavior 1. Argues a lot 2. Cruelty, bullying, or meanness to others 3. Demands a lot of attention 4. Destroys his/her own things 5. Destroys things belonging to his/her family or others 6. Disobedient at home 7. Disobedient at school 8. Gets in many fights 9. Physically attacks people 10. Screams a lot 11. Stubborn, sullen, or irritable 12. Sudden changes in mood or feelings 13. Sulks a lot 14. Suspicious
74 Appendix B (Continued) 15. Teases a lot 16. Temper tantrums or hot temper 17. Threatens people 18. Unusually loud Internalizing Subscales: Withdrawn/Depressed 1. There is very little he/she enjoys 2. Would rather be alone than with others 3. Refuses to talk 4. Secretive, keeps things to self 5. Too shy or timid 6. Underactive, slow moving, lacks energy 7. Unhappy, sad, or depressed 8. Withdrawn, doesnÂ’t get involved with others Anxious/Depressed 1. Cries a lot 2. Fears certain animals, situati ons, or places, other than school (describe): _____________________ 3. Fears going to school 4. Fears he/she might think or do something bad 5. Feels he/she has to be perfect 6. Feels or complains that no one loves him/her 7. Feels worthless or inferior 8. Nervous, highstrung, or tense 9. Too fearful or anxious 10. Feels too guilty 11. Self-conscious or easily embarrassed 12. Talks about killing self 13. Worries
75 Appendix C OÂ’LEARY-PORTER SCALE R EVISED VERSION: Parents Please answer all of the following questions to the best of your ability. The questions refer to your son/daughter, only. PLEASE NOTE: The term Â“spouseÂ” refers to your sonÂ’s/daughterÂ’s other parent, regardless of w hether you are currently married to him or her. 1. It is difficult in these days of tight budgets to confine financial discussions to specific times and places. How often would you say you and your s pouse argue over money matters in front of this child? Never Rarely Occasionally Often Very Often 2. Children often go to one parent for money or permission to do something after having been refused by the other parent. How often would you say this child approaches you or your spouse in this manner with rewarding results? Never Rarely Occasionally Often Very Often 3. Husbands and wives often disagree on the subj ect of discipline. How often do you and your spouse argue over discipline problems in this childÂ’s presence? Never Rarely Occasionally Often Very Often 4. How often has this child heard you and your spouse argue about the wifeÂ’s role in the family? (Hours of work, mothering behaviors, etc.) Never Rarely Occasionally Often Very Often 5. How often does your spouse complain to y ou about your personal habits? (drinking, nagging, sloppiness, etc.) in front of this child? Never Rarely Occasionally Often Very Often 6. How often do you complain to your spouse about his/her personal habits in front of this child? Never Rarely Occasionally Often Very Often 7. In every normal marriage there are argum ents. What percentage of the arguments between you and your spouse would you say take place in front of this child? Less than 10% 10-25% 26-50% 51-75% More than 75% 8. To varying degrees, we all experience almost ir resistible impulses in time of great stress. How often is there physical expres sion of hostility between you and your spouse in front of this child? Never Rarely Occasionally Often Very Often 9. How often do you and/or your spouse display verbal hostility in front of this child? Never Rarely Occasionally Often Very Often 10. How often do you and your spouse display affection for each other in front of this child? Never Rarely Occasionally Often Very Often 11. How often has this child heard you and your spouse argue about the husbandÂ’s role in the family? (Hours of work, fathering behaviors, etc.) Never Rarely Occasionally Often Very Often
76 Appendix D PARENTAL DEMOGRAPHIC FORM Please complete the following: 4. This form is being completed by a: Mother ___ Stepmother ___ Adoptive mother ___ Father ___ Stepfather ___ Adoptive father ___ Guardian ___ Other ___ 2. How old are you? ___ 3. What is your race /ethnicity? ______________________ 4. How many children (biological, stepc hildren, and other children) are presently living in your home? ___ 5. List the ages of t he children who are presently living in your home: __________________ __________________ 6. In all, how many children (biologica l, stepchildren, and others) do you have? ________ 7. How many siblings did you have growing up? ___ 8. Were you the oldest, y oungest, or middle child? ______ 9. Are you: ___Married ___Single, living wit h partner ___Singl e, no partner ___Separated ___Divorce d ___Widowed ___Other 10. Your employment status: Mother or Female Guardian Father or Male Guardian (either you or your partner) (either you or your partner) Employed as _______________ Employed as _______________ Unemployed _______________ U nemployed _______________ Retired _________________ __ Retired ___________________ Other ____________________ Other ____________________
77 Appendix D (C ontinued) 11. What percentage of your professi onal involvement is spent working with children (0-100%)? ______ 12. On a scale of 1-10, where 1=Not at all and 10=A lot, how much experience have you had with children other than your own (either in a work or personal capacity)? ______ 13. Highest education level completed: Mother/Female GuardianÂ—Year s of Education: _______________ Father/Male GuardianÂ—Years of Education: __________________ 14. Total household income per year : _______________ 15. Average hours per week you spend at work and/or school, including commuting time? ______ 16. In an average week day, how much time do you spend with your child(ren) during waking hours? ______ 17. In an average weekend day, how much time do you spend with your child(ren) during waking hours? ______ 18. Has either of your child(ren)Â’s parents received mental health services (such as therapy, counseling, or medication) in order to deal with something that was psychologically distressing? ______ Yes _______ No If Yes: Please note who received the services, what type of services were received (e.g., psychiatrist, pastoral c ounseling, etc.), and how long ago the services were received. Please use back of page if you need additional space. ________________________ _____________________ ___________________ 19. Have any of your children received me ntal health services in order to deal with something that was psychologi cally distressing? ______ Yes ______ No If Yes: Please note who received the services, what type of services were received, and how long ago the services were received. ________________________ __________________ ______________________ ________________________ ___________________________ _____________
78 Appendix E Questionnaire for Raters 1. Your Name: _________________________ 2. Your Gender: 1. Male 2. Female 3. Your Race: 1. African American 2. Caucasian 3. Hispanic 4. Asian 5. Other (please specify): _______________________ 4. Your year in Grad school (put Â“N/AÂ” if Faculty): ____________ Which program? 1. Clinical 2. I/O 3. CNS 4. Other (please specify): _________________ If Clinical how many years of experience have you had with child clients? ____ Please explain briefly: ______________________________________________ ________________________________________________________________ 5. Year as faculty (put Â“N/AÂ” if grad student): _____________ Which program? 1. Clinical 2. I/O 3. CNS 4. Other (please specify): ________________ If Clinical how many years of experience have you had with child clients? ____ Please explain briefly: ______________________________________________ ________________________________________________________________ 6. How many classes have you taken on child psychopathology/development? ____ 7. How many classes have you taught on child psychopathology/development? ____ 8. How many years of professional experience (including paid and volunteer) have you had working with children? _________ Please explain briefly:
79 Appendix E (Continued) Girl Child Actor Video Instructions: Please select the behavior and the intensity of t he behavior that the child actor displays. Note that you wil l watch three video segments with a girl child actor, one of which displays internalizing behavior, one of which displays externalizing behavior, and one of which di splays non-clinical behavior. Thus, this is a forced-choice design in which you can only choose each answer once (i.e., internalizing, exter nalizing, or non-clinical). Segment 1 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense Segment 2 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense Segment 3 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at All Somewhat intense Very intense
80 Appendix E (Continued) Boy Child Actor Video Instructions: Please select the behavior and the intensity of t he behavior that the child actor displays. Note that you will watch three video segments with a boy child actor, one of which displays internalizing behavior, one of which displays externalizing behavior, and one of which di splays non-clinical behavior. Thus, this is a forced-choice design in which you can only choose each answer once (i.e., internalizing, exter nalizing, or non-clinical). Segment 1 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense Segment 2 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense Segment 3 Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense
81 Appendix F Additional Rating Form Name: ____________________ Boy Child Actor Video Instructions: You will watch one video segm ent with a boy child actor. Please indicate if the child displays internalizi ng, externalizing, or non-clinical behavior and the intensity of his behavior. Child displays which of th e following (please circle): Internalizing Behavior Externalizi ng Behavior Non-Clinical Behavior Degree of Intensity of Behavior displayed (please circle): 1 2 3 4 5 Not intense at all Somewhat intense Very intense
82 Appendix G Space below reserved for IRB Stamp Â– Please leave blank Informed Consent Social and Behavioral Sciences University of South Florida Information for People Who T ake Part in Re search Studies The following information is being present ed to help you decide whether or not you want to take part in a minimal risk res earch study. Please read this carefully. If you do not understand anything, ask t he person in charge of the study. Title of Study: Ratings of ChildrenÂ’s Behavior Principal Investigator: Jessica K. Curley Study Location(s): University of South Flori da Psychology Department You are being asked to participate because you are a parent of a child between the ages of 4 and 21. General Information about the Research Study The purpose of this research study is to better understand fact ors that contribute to parental ratings of childrenÂ’s behavior. Plan of Study You will be asked to do the following: Wa tch an 8 minute video of a child being interviewed and engaging in free play. You will then be asked to rate the childÂ’s behavior on several dimensions. You will also be asked to fill out a demographics questionnaire and 2 questionnaires relating to your functioning. The entire study should take about 30 minutes. Payment for Participation You will not be paid for your participation in this study. However, if you are a psychology student from the USF Partici pant Pool, you will receive extra credit points towards a psychology course for y our participation. If you are not a psychology student, or if you do not want the extra credit points, you will be entered into a drawing for one of two $50 prizes or one $100 prize. Benefits of Being a Part of this Research Study By taking part in this research study, y ou may increase your overall knowledge of how childrenÂ’s behavior is viewed. You will also be contributing to the understanding of factors that influenc e ratings of childrenÂ’s behavior.
83 Appendix G (Continued) Risks of Being a Part of this Research Study This study should pose no physical or psychological harm to you. The questionnaires may result in minimal levels of distress in that they ask you about potentially troubling behaviors, emotions, and events. However, all measures have been standardized and utili zed previously in research settings with no known adverse effects. Confidentiality of Your Records Your privacy and research records will be k ept confidential to the extent of the law. Authorized research personnel, empl oyees of the Department of Health and Human Services, and the USF Instituti onal Review Board may inspect the records from this research project. The results of this study may be published. However, the data obtained from you will be combined with data from others in the publication. The published results will not include your name or any other information that would personally identify you in any way. All records will be identified by numbers and your identity will not be placed on any of the completed forms. Access to the data will be restricted to relevant students and faculty of the Psychology Depar tment at the University of South Florida. Volunteering to Be Part of this Research Study Your decision to participate in this resear ch study is completely voluntary. You are free to participate in this research study or to withdr aw at any time. There will be no penalty or loss of benefits you are entitled to receive, if you stop taking part in the study. Questions and Contacts If you have any questions about this research study, please contact Jessica K. Curley, Department of Psychology, Univer sity of South Florida, 4202 E. Fowler Ave. PCD 4118G, Tampa, FL 33620, 813-974-9222, firstname.lastname@example.org If you have questions about your rights as a person who is taking part in a research study, you may contact the Division of Research Compliance of the University of South Florida at (813) 974-5638. Consent to Take Part in This Research Study By signing this form I agree that: I have fully read or have had read and explained to me this informed consent form describing this research project.
84 Appendix G (Continued) I have had the opportunity to question one of the persons in charge of this research and have received satisfactory answers. I understand that I am being asked 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 Participant Print ed Name of Participant Date Investigator Statement I have carefully explained to the subject the nature of the above research study. I hereby certify that to t he best of my knowledge the s ubject signing this consent form understands the nature, demands, risks, and benefits involved in participating in this study. _________________________ _________ _______________ _________ Signature of Investigator Print ed Name of Investigator Date Or authorized research investigator designated by the Principal Investigator
85 Appendix H Debriefing Form The goal of this study was to examine how moth ers and fathers view childrenÂ’s behavior. More specifically, we wanted to see how certai n factors, such as parentsÂ’ experiences with children, parentsÂ’ functioning, and certain charac teristics of childrenÂ’s behaviors are related to mothersÂ’ and fathersÂ’ disagreement on ratings of childrenÂ’s behavior. Previous research has not examined parental ratings of ch ildrenÂ’s behavior using a videotape with a child actor like the one that you just viewed. Therefore, in this study, we controlled for your knowledge of the child that you were rating and had different parents rate different videotapes. The knowledge to be gained by this research will include identifying other factors, beyond those controlled for, which influence parental disagreement about child behavior. When childrenÂ’s behavior is problematic, parents may want to have a clinician address such issues as better family relations and proper treatment for a child. Custodial parents are most commonly the ones who refer children for treatment. If the parents disagree on the problems a child is exhibiting, then that child may be restricted from receiving proper treatment. Focusing on factors that are associated with disagreement can lead to prevention efforts that will allow for parents to engage their child in treatment before the problems become too severe. If you would like to learn more about parental ratings of child behavior problems, here are three journal articles to consult: Achenbach, T.M., McConaughy, S.H., & Howell, C. T. (1987). Child/adolescent behavioral and emotional problems: Implications of cross-inform ant correlations for situational specificity. Psychological Bulletin, 101, 213-222. Duhig, A., Renk, K., Epstein, M., & Phares, V. ( 2000). Interparental agreement on internalizing, externalizing, and total behavior problems: A meta-analysis. Clinical Psychology: Science and Practice, 7, 435-453. Phares, V., Compas, B.E., & Howell, D.C. (1989) Perspectives on child behavior problems: Comparisons of childrenÂ’s self-reports with parent and teacher reports. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 68-71. If you or someone you know is concerned about t heir childÂ’s behavior or their own behavior, here are some resources to consider: USF Counseling Center (for USF students who are seeking help for themselvesÂ—No cost to students): 813-974-2831 USF Psychological Services Center (for students and their families as well as for individuals from the communityÂ—small fee on sliding scale based on ability to pay): 813-974-2496 Northside Community Mental Health Center (for individuals from the community who are seeking helpÂ—sliding scale based on ability to pay): 813-977-8700 Some insurance companies also cover mental health services, so please feel free to check with your health insurance company to see if they can cover psychological evaluations or treatments if you are interested. If you have any questions or concer ns regarding this study, please contact Jessica Curley at 813-974-9222 or email, email@example.com Thank you for your participation!!