..-—- -, o A -';:J'.','. . .r 11:23:...‘3, :x n; ‘05.; , . :-L3‘. 2 —‘xx{& 73. >l~m?' . \; u. - 1 «v.6»...“l‘ Wu ‘ «.1314‘ ,“_ ‘ Iti"'w‘\}3lj”~: Sank-Jul." ”A“ F“? ‘1. 1, ”$133.7 3552;: - ;.;;);:A.~,m I." . $1-41: . w .1‘fi’ll-. 9: 13. 1., “Mn-”a; .4,¢.:..4. '4, -~ “ \‘ r-E‘N “p.3- .vfl-Ec‘ 9313‘” "“- f»! . ~ 33%;; .«r I-i‘O-l’ .34 5:4,... S5 .. . 55—: nbr : My, than.-. , , .. - ‘ 54‘9"- ”mi-33321.3-.. - ”'53....94..-. cw:- ""v "5' ‘0‘].— ‘ by “on , 4-55»; my. 5mm. V "‘"r'fl' lllllllllllllllllllllllllllllllllllllllllllllllllllllll 3 1293 00911 691 This is to certify that the thesis entitled Victim Blame and Child Abuse presented by Robert Tom Muller has been accepted towards fulfillment of the requirements for Masters Psychology degree in mafia / @447 Major professor Date A I" 1 991 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution F LIBRARY Michigan State \ University ,l PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE W E; —T _l IT—ll l l—l l__ l MSU In An Affinnative Action/Equal Opportunity Institution cmmut VICTIM BLAME AND CHILD ABUSE BY Robert Tom Muller A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1991 (RTQwafllfijfl ABSTRACT VICTIM BLAME AND CHILD ABUSE BY Robert Tom Muller The current research was an investigation of victim blame in the case of child physical abuse. The subjects were 866 college undergraduates. The research consisted of three studies. In the first study, several situational determinants of victim blame in child abuse were investigated. The results suggested that the situations leading to greater degrees of child blame were those in which the victims were provocative children, male children, and in which the abusers were male parents. In the second study, individual differences in victim blame were investigated. The results were consistent with the position that victim blame is a global personality factor. Persons who blamed victims in one domain of conflict, did so in others as well. The third study was an investigation of the process by which individuals may go from having experienced physical abuse as a child to subsequent child blame. The results suggested that the prior experience of abuse predicts changes: (a) in the individual's view of the self; (b) in the way the person relates to others; and (c) in the person's assumptions about the nature of his/her environment. These factors appear to converge in predicting greater blame of abused children. To my bride, Diane ACKNOWLEDGEMENTS I would like to express my profound gratitude to my committee chairperson, Dr. Robert A. Caldwell. In the unpredictable cacophony of graduate school, he has provided me with an atmosphere of consistency and stability. His high degree of integrity and trustworthiness have been a saving grace. Whether the research problem was large or small, he has shared with me willingly his wisdom, support, and encouragement. At the same time, he has allowed me the space to exercise my creative potential. For this too, I am grateful. I would like to declare my deepest appreciation to Dr. John E. Hunter for serving on my committee. Dr. Hunter has .been a key player in this project from its inception. By challenging my theoretical and methodological formulations, he has shaped my thinking about research. I have treasured our many debates on politics, economics, history, and of course, psychology. He has also provided me with much support in times of personal distress. Finally, I would like to acknowledge my sincere gratitude to Dr. Gary Stollak, who also served on my committee. He has been eager to share with me his clinical wisdom and expertise. I have appreciated greatly his iv insights and conceptualizations regarding the nature of child maltreatment. Dr. Stollak's sense of humor and supportive posture have been invaluable ingredients in the completion of this project. Finally, I would like to thank several undergraduate students for their assistance with data management and coding. These are Robert Tinney, Annie Gurley, and Adrian Samuels. TABLE OF CONTENTS Page LIST OF TABLESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOx LIST OF FIGURESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO x11 INTRODUflIONOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Study I: Factors Contributing to Child Blame..... Gender of the Subject........................ Provocativeness of the Child................. Gender of the Abuser......................... Gender of the Child.......................... Hypotheses l-4............................... mmumosox H Study II: The Construct Dimensionality of Victim BlameO O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 10 Hypotheses 5.6 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 11 Study III: The Path from Survivor of Abuse to Child Blame................................... 12 Intervening Variables........................ l4 Empathy................................. l4 Belief in a Just World.................. 15 Locus of Control........................ 17 Personal Similarity..................... 19 Self-Derogation......................... 21 HYPOtheses 7-18eeeeeeeoeeeeoeeeoeoeeeeeeeeeee 23 Overall Relationship between Abuse as a Child and Child Blame....................... 25 SOCIO"eCOnOlniC StatuSeeeeoeeooeeeeeoeeeeeeeoo 26 Additional Issues Explored: Gender Concordance on Personal SimilaritYO O O O O O O O O O O O O O O O O O O O O O O O O O O 27 HYPOth6$1s 19O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 28 WODOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 29 subjectSO O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 29 MateriaISO O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 30 procedureOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 40 vi Page RESULTsmDISCUSSIONOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 43 Provocativeness manipulation check................ 43 study I: Factors Contributing to Child Blame; Results.......................................... 45 Hypothesis 1................................. 48 Hypothesis 2................................. 48 Hypothesis 3................................. 48 Hypothesis 4................................. 49 Treatment Interaction Effects................ 49 Provocativeness by Abuser Gender........ 49 Provocativeness by Child Gender......... 50 Abuser Gender by Child Gender........... 50 Treatment by Subjects Interaction Effects.... 50 Study I: Factors Contributing to Child Blame; Discussion....................................... 52 Hypothesis 1 (Gender of the Subject)......... 52 Hypothesis 2 (Provocativeness of the Child)...................................... 52 Hypothesis 3 (Gender of the Abuser).......... 54 Hypothesis 4 (Gender of the Child)........... 55 Study II: The Construct Dimensionality of Victim Blame: Results............................ 58 Confirmatory Factor Analyses Results......... 59 Conflict Tactics Scales................. 59 The Interpersonal Reactivity Index...... 60 The Just World Scale.................... 60 The I, P, and C Scales.................. 61 The Self-Derogation Index............... 61 The Child Abuse Scenarios Questionnaire.......................... 61 Test of Hypothesis 5.................... 69 The Attitudes Toward Rape Victims Scale.......................... 70 The CASQ and the ARVS Together: A Test for a Global Victim Blame Construct (Test of Hypothesis 6)....... 77 Reliabilities................................ 79 Scale Means, Standard Errors, and Standard Deviations.................................. 79 Study II: The Construct Dimensionality of Victim Blame; Discussion......................... 82 HYPOtheSis SOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 82 Hypothesis 6................................. 85 vii Study III: The Path from Survivor of Abuse to Child Blame: Results.......................... Overall Relationship between Abuse as a Child and Child BlameOOOOOOOOOOOOOOOOOOOOOOO Test of Hypotheses 7-18...................... Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis 7............................ 8000OOOOOOOOOOOOOOOOOOOOOOOOO 9000OOOOOOOOOOOOOOOOOOOOOOOOO loOOOOOOOOOOOOOOOOOOOOOOOOOOO 11000OOOOOOOOOOOOOOOOOOOOOOOO 12........................... 13........................... 14000OOOOOOOOOOOOOOOOOOOOOOOO 15........................... 16........................... 17........................... 18000OOOOOOOOOOOOOOOOOOOOOOOO The Path AnalyseSOOOOOOOOOOOOOOOOOOOOOOOOOOOO Study III: Additional Issues Explored: The Path from Survivor of Abuse to Child Blame; Discussion....................... Overall Relationship between Abuse as a Child and Child Blame....................... Empathy................................. Just World Beliefs...................... 98 Locus of Control........................100 Personal Similarity.....................103 Self-Derogation.........................106 Summary of Study III.........................109 Methodological Limitations...................lll The Effects of Gender Page 88 88 88 88 89 89 89 89 9O 9O 9O 90 91 91 91 92 95 95 96 Concordance on Personal Similarity: Results......113 HYPOtheSis 19000OOOOOOOOOOOOOOOOOOOOOOOOOOOOO113 Additional Issues Explored: The Effects of Gender Concordance on Personal Similarity: Discussion...ll4 HYPOtheSis 19000OOOOOOOOOOOOOOOOOOOOOOOOOOOOO114 POSt-Hoc Findings; ResultSOOOOOOOOOOOOOOOOOOOOOOOOllS Post-Hoe Findings; Discussion.....................117 6mm DISCUSSIONOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO119 Methodological Limitations of all Three Studies...122 viii Page APPENDICESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO124 Conflict Tactics Scales...........................126 Self-Derogation Index.............................129 Child Abuse Scenarios Questionnaire...............131 Duncan Socio-economic Index.......................149 Instructions/Letter of Consent....................152 Instructions for Second Administration............154 Feedback Form.....................................156 LIST OF REFERENCESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO158 ix LIST OF TABLES Table Page 1. Analysis of Variance of Child Blame: Subject Gender by Child Provocativeness by Abuser Gender by Child Gender............................ 46 2. Group Means and Standard Deviations of Child Blame Broken Down by Provocativeness, Abuser Gender, and Child Gender (Female Subjects Only)... 47 3. Group Means and Standard Deviations of Child Blame Broken Down by Provocativeness, Abuser Gender, and Child Gender (Male Subjects Only)..... 47 4. Items from the Behavioral Child Blame Scale and their Factor Loadings onto the three Dimensions of Child Blame......................... 62 5. Items from the Characterological Child Blame Scale and their Factor Loadings onto the three Dimensions of Child Blame......................... 65 6. Items from the Abuser Justification Scale and their Factor Loadings onto the three Dimensions of Child Blame......................... 68 7. Correlations of the three Dimensions of Child Blame with Outside Constructs............... 72 8. Items from the Rape Victim Blame Scales and their Factor Loadings onto the four Dimensions of Rape Victim Blame................... 73 9. Correlations of the four Dimensions of Rape Victim Blame with Outside Constructs......... 76 lo. Correlations of the two Dimensions of Victim Blame with Outside Constructs.............. 78 11. Parallel-Forms and Test-Retest Reliability EstimateSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 80 12. Scale Means, Standard Errors, and Standard Deviations........................... 81 X Table Page 13. Factor by Factor Correlation Matrix...............88A xi LIST OF FIGURES Figure Page 1. A Proposed Path from Childhood Experience of 2. Physical Abuse to Child Blame.....................12A Scatterplot of the Relationship between Child Blame Scores for Provocative and Nonprovocative Scenarios..........................51A Higher Order and Lower Order Victim Blame Factors and their Factor Loadings.................69A The Proposed Path Model Adjusted to Reflect the Unidimensionality of Constructs...............93A The Path Model with Correlations and Path Coefficients between Constructs..............938 The Path Model that Provided Best Fit to the DataOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO93C The Effect of Physical Abuse and Siblings on ChildhOOd Self-Derogatione O O O O O O O O O O O O O O O O O O O O O 115A The Effects of Physical Abuse, Having no Siblings, and their Interaction on Childhood Self-Derogation.........................IISB xii INTRODUCTION A considerable literature has developed on the tendency to ascribe the "blame" for unfortunate life events on the victims of the misfortune. Ryan (1971) suggested that, with respect to social problems, there is a tendency in North American society to localize both the source of the problem as well as the solution in those who are suffering. He referred to this tendency as "blaming the victim". The poor, for example, are commonly seen as unmotivated rather than being seen as restricted in terms of opportunity. The tendency to blame victims has been addressed by several theories in social psychology. Lerner's (1980) ”just world" theory asserted that people generally believe that the world is just and fair. Instances in which good things happen to bad people or good people suffer, threatens the conception of a just world. In many cases, individuals respond to such inequities by altering their perception of the victim or the victim's behavior so that the victim is devalued and blamed for the misfortune (alternate ways of restoring justice in Lerner's scheme are compensating the victim and/or punishing the tormentor). Considerable empirical evidence has been obtained consistent 2 with the just world hypothesis (cf. Kerr 8 Kurtz, 1977: R. E. Smith, Heating, Hester, & Mitchell, 1976). Another paradigm used to explain victim blame is that of locus of control. Internal-external control is an individual difference variable which deals with the extent to which persons attribute responsibility for the occurrence of reinforcement to themselves (internals) or to forces outside themselves (externals). Phares and Lamiell (1975) suggested that internal persons view themselves as generally responsible in their lives. Similarly, internals view others as responsible for their goal attainment or lack of it. Several studies have indicated that internals are more likely than externals to attribute fault to victims of rape (Paulsen, 1979), and to victims of auto accidents (Sosis, 1974). Related to the phenomenon of victim blame is the tendency to blame one's self, following one's own misfortune. This phenomenon has been found in rape victims (e.g., Ledray, 1986: Libow & Doty, 1979). The most common explanation for self-blame is that victims feel uncomfortable with the belief that they are helpless or vulnerable individuals. It is comforting to feel that one could have somehow prevented the disaster. Blaming one's self restores a sense of control of outcome (Symonds, 1975). Janoff-Bulman (1979) distinguished between two kinds of self-blame: behavioral and characterological. Behavioral 3 self-blame refers to attributions of fault with respect to one's behavior. In other words, individuals may blame themselves for having engaged in a certain activity. In blaming one's behavior, the individual is concerned with the future avoidability of the negative outcome. Behavioral self-blame is control motivated. Characterological self- blame refers to attributions of fault concerning one's character. An individual may blame him/herself for the kind of person s/he is. Characterological self-blame is esteem related. Janoff-Bulman (1979) suggested that behavioral self-blame is typical of rape victims and characterological self-blame is typical of depressed persons. Applying Janoff-Bulman's self-blame distinction to victim blame, Thornton (1984) suggested that when evaluating "personally similar" victims (i.e., victims similar to themselves), observers will invoke behavioral blame, determining that something the victim did (or failed to do) brought about the consequence. Due to the perception that behavior is controllable, observers may make such attributions to a similar victim with the belief that they themselves can behave in a different manner and avoid negative consequence. Moreover, evaluating personally similar victims, observers will be less inclined to ascribe characterological blame as that would depict one's own stable character in a negative light. 4 Related to the tendency to ascribe blame to victims is the notion that some individuals, under certain circumstances, may tend to justify the actions of violent offenders (justifying the perpetrator). Libow and Doty (1979) discovered many severely injured rape victims who felt sorry for their assailants and recommended only lenient treatment. Krulewitz and Nash (1979) suggested that rapists who "complete” an attack are commonly seen as less responsible, since the victim is perceived as not having resisted the attacker adequately. Jenkins and Dambrot (1987) found that sexually aggressive males, and males who agreed more with rape myths viewed the behaviors of assailants as less violent. Krahe (1988) determined that individuals who scored high on rape myth acceptance ascribed less responsibility to rapists. Herzberger and Tennen (1985) found that, in reacting to scenarios of abusive parental discipline, subjects rated fathers' identical behaviors as more appropriate than those of mothers. Cohen (1984) suggested that abused children will justify their parents' behaviors as an act of familial "loyalty". Despite the substantial body of literature dealing with the circumstances surrounding victim blame and perpetrator justification, research heretofore had neglected to explore systematically these phenomena in the case of physical child abuse. The present study addressed this issue. Specifically, the primary purpose of this study was to 5 examine the effects of several seemingly relevant factors on child blame. Child blame is the tendency to hold abused children responsible for their misfortune, and it is divided into three components: Behavioral, characterological, and abuser justification. Behavioral and characterological facets of child blame refer to the tendency to view the child as responsible because of its behavior (actions) or character (personality). Abuser justification refers to the tendency to view abusive parental discipline as justifiable. 6 Study I WWII: ggng§;_gfi_;hg_§nhjggg. Several investigations have demonstrated gender differences in the attribution of blame. Males seem to attribute more blame to victims than do females. This finding has been demonstrated using situations of rape (Calhoun, Selby, Cann, & Keller, 1978: Deitz, Blackwell, Daley, & Bentley, 1982; Kanekar & Kolsawalla, 1980), child molestation (Waterman & Foss- Goodman, 1984), and child physical abuse (Herzberger & Tennen, 1985). It was suggested that in the current study as well, males would attribute greater child blame. 2rggagatiygn§§§_gf_tng_gn11d. Several studies on the causes of physical child abuse have indicated that certain types of children are more likely candidates for abuse than others. Reid, Patterson, and Loeber (1981) reported that children difficult to manage (e.g., hyperactive children) are more likely to be physically abused. J. E. Smith (1984) considered verbally aggressive children to be at high risk for abuse. Engfer and Schneewind (1982) as well as Herrenkohl, Herrenkohl, and Egolf (1983) noted that children perceived to be difficult to handle are candidates for abuse. In general, it appears that apparently provocative children are more often victims of physical abuse. The question remains, however, as to whether individuals see such children as responsible and whether they see such 7 parental behavior as justifiable with troublesome children. The recent child abuse literature indicates increasing concern with the role of cognitive factors in child maltreatment (Larrance & Twentyman, 1983). Cognitive perceptions, interpretations and attributions have been proposed as an antecedent of physical abuse (Bakan, 1971: Helfer, McKinney & Kempe, 1976). Several studies (e.g., Larrance & Twentyman, 1983: Rosenberg & Reppucci, 1983) have been conducted to investigate differences between abusive and nonabusive parents in attributing responsibility to their own provocative children. However, it appears that prior researchers have neglected to investigate attributions of responsibility to provocative children in_ggmparisgn_with non-provocative children. In other words, is the provocativeness itself a factor in the elicitation of attributions of responsibility? If so, this would imply the existence of mediating cognitions and attitudes in the greater abuse of provocative children. The current study investigated the possibility that child provocativeness may lead to greater child blame. fiendgz_gf_the_bbn§§:. Another seemingly relevant factor in the elicitation of child blame is the gender of the abuser. Herzberger and Tennen (1985) demonstrated that, in reacting to vignettes describing the administration of abusive parental punishment, subjects rated mothers’ identical behaviors as less appropriate than those of 8 fathers. The results were discussed in terms of the stereotyped expectation that mothers should be nurturant and loving. The current study investigated the relationship between gender of the abuser and abuser justification. It was suggested that male abusers would elicit greater sentiments of abuser justification. Gendez_gf_thg_ghild. Studies focused on the investigation of gender differences in child abuse have demonstrated, in general, that male children are somewhat more likely to be abused than their female counterparts (e.g., Gil, 1970: Lieh-Mak, Chung, 8 Liu, 1983; Maden 8 Wrench, 1977). In judging harshness of parental punishment, discipline of daughters was judged to be less appropriate than identical treatment of sons (Herzberger 8 Tennen, 1985). In one study of child sexual abuse, Waterman and Foss-Goodman (1984) found a tendency for male children to be blamed more than female children. Males may be blamed more due to a cultural stereotype of innocence of girls as compared to boys. The current investigation examined the effect of the child’s gender on the tendency to child blame. It was proposed that male children would elicit greater child blame. W13 1. Male subjects, in comparison to females, should ascribe more child blame. 9 2. When rating provocative, in comparison to nonprovocative children, subjects should ascribe greater child blame. 3. When rating male, in comparison to female abusers, subjects should indicate more abuser justification. 4. When rating male, in comparison to female children, subjects should indicate higher levels of child blame. 10 Study II WWW One purpose to this study was to establish the dimensionality of the victim blame construct. Although many studies have demonstrated the tendency to blame victims of personal misfortune (Lerner, 1980: Ledray, 1986), few investigations have been conducted to determine the dimensionality of the victim blame construct. As discussed previously, Janoff-Bulman (1979) distinguished between behavioral and characterological components of victim blame and concluded that behavioral self-blame is typical of rape victims, while characterological self-blame is typical of depressed persons. The Janoff-Bulman (1979) distinction has influenced the development of many research programs (e.g., Jensen 8 Gutek, 1982: Meyer 8 Taylor, 1986: Schoeneman, Hollis, Stevens, 8 Fischer, 1988: Thornton, 1984), as well as theoretical analyses (e.g., Shaver 8 Drown, 1986). For example, Thornton (1984) found that in evaluating personally similar victims, behavioral blame was ascribed more than characterological blame while the reverse was true in evaluating personally dissimilar victims. Prior research has also assumed that victim blame in the case of rape is separate and distinct from victim blame in other areas, such as child abuse. Several scales have been developed which assume specificity of these constructs. 11 For example, Ward's (1988) Attitudes Toward Rape Victims Scale (ARVS) was designed to measure unfavorable or unsupportive attitudes toward rape victims. Similarly, Deitz and colleagues developed an empathy scale specific to the rape situation (Deitz, Blackwell, Daley, 8 Bentley, 1982: Deitz, Littman, 8 Bentley, 1984). One of the purposes of the current study was to investigate the dimensionality of victim blame. Based on prior research, it was suggested that victim blame would be a multidimensional construct. More specifically, the following was hypothesized: W 5. Behavioral and characterological components of victim blame, as well as abuser justification would all emerge as distinct factors. 6. Blame directed toward rape victims and blame directed toward victims of child abuse would also emerge as separate and distinguishable constructs. 12 Study III Wane. In investigating child blame, one factor which may be relevant is that of history of childhood physical abuse. The potential importance of investigating the relationship between being an abused child and blaming abused children for their misfortune is explained as follows. Several researchers have pointed out a common finding in the child abuse literature, the high concordance between being a survivor of physical child abuse and being a perpetrator of child abuse (e.g., Carroll, 1977: Gillespie, Seaberg, 8 Berlin, 1977: Isaacs, 1981; Lieh-Mak, Chung, 8 Liu, 1983: Webster-Stratton, 1985), the so-called "cycle of abuse" (Kempe 8 Kempe, 1978). Simons, Whitbeck, Conger, and Chyi- In (1991) provided validation of a social learning model for the intergenerational transmission of harsh parenting. If the cycle of abuse is, indeed, a true phenomenon, and if adult survivors of abuse blame abused children to a considerable extent, we may infer that the resulting child abuse situation is experienced as ego-syntonic in the mind of the abuser. In other words, if survivors of abuse cognitively structure their environments so as to blame the child and justify the abuser, their subsequent abusive behaviors may be experienced as consonant with the self. Alternatively, if adult survivors of abuse blame abused children much less than controls, their subsequent abusive Egsgahgmhgnfimgfiofifge 33.3 153.3 BIB...» 886118938 \ + Susana: _ ~ 8.38.8...» 818.8...» 32.8 All Bea-6 + + + ‘\ - is .. \5’09 .9 afipr.’ H 2...... 3'9”” 4v 3.... - I g .5338 u .588 l3 behaviors would be experienced as dissonant with the self (ego-dystonic). If these two approaches both prove to be relevant, this would suggest that the act of child abuse may be experienced very differently for various individuals. Furthermore, it would appear that those individuals who maltreat their own children and who perceive the child as responsible would be much more resistant to change and less apt to benefit from therapeutic intervention. It is instructive to note that some investigators consider the "cycle of abuse" to be overstated. Kaufman and Zigler (1987) noted that only one third of adults who were abused become abusers themselves. Furthermore, Gil (1970) indicated that only 14% of the mothers and 7% of the fathers in abuse cases had been victims of abuse in their childhood. As such, it appears that the path from abusive childhood to child abuser is neither simple nor direct. The theoretical conceptualizations presented so far only provide a possible framework for explaining the cognitive processes involved in going from abused child to child abuser. They do not provide an explanation for the many cases which do not involve the intergenerational component. Prior research allowed for the inference of a possible path from childhood experience with physical abuse to child blame. This path is presented as Figure l. 14 MW Empathy; The construct of empathy can be delineated into two broad classes of response: a cognitive reaction (an ability to understand the other person’s perspective), and a more visceral, emotional reaction. Empathy ~conceptualized as a cognitive response has focused on processes such as accurate perceptions of others (e.g., Dymond, 1949: Kerr & Speroff, 1954), while researchers concentrating on the emotional aspects of empathy (see Mehrabian 8 Epstein, 1972; Stotland, Mathews, Sherman, Hansson & Richardson, 1978) have focused on such issues as "emotional reactivity" and helping behavior. Recently, several theorists (e.g., Davis, 1980, 1983: Hoffman, 1977) have suggested that empathy is a multidimensional construct consisting of both cognitive and emotional components. Earlier research has shown that abused children score significantly lower on measures of empathy than nonabused children (Main 8 George, 1985; Miller & Eisenberg, 1988; Straker & Jacobson, 1981). In a longitudinal investigation, Koestner, Franz, and Weinberger (1990) found that various nurturant parental behaviors directed toward children aged 5, predicted empathic behaviors in those individuals at age 31. In a study investigating the consequences of child abuse, Straker and Jacobson (1981) found abused children to have significantly lower levels of empathy than nonabused children. However, Straker and Jacobson failed to 15 differentiate between cognitive and emotional facets of empathy. In Figure 1, it was suggested that as adults, survivors of abuse should demonstrate low levels of empathy (both cognitive and emotional). Prior investigators have studied the extent to which observer empathy influences attributions of responsibility. Deitz, Blackwell, Daley, and Bentley (1982) and Deitz, Littman, and Bentley (1984) found that individuals demonstrating low levels of rape empathy were more likely to blame victims of rape. In contrast, Collar and Resick (1987) did not find a relationship between these two variables. However, in the Coller and Resick study, observer empathy was "manipulated," rather than treated as an individual difference variable. The potential success of such a manipulation appears somewhat questionable. Thus, in the current study, empathy was treated as an individual difference variable. In Figure 1, it was suggested that individuals demonstrating low levels of each of the two types of empathy (cognitive and emotional) should be more likely to child blame. Beligf_in_a_qust_flgzld‘ In Lerner’s (1980) view, it is difficult for individuals to accept the arbitrariness which characterizes the world. He suggested that people need to believe that they live in a world in which they get what they deserve. The existence of good people suffering threatens this conception. As such, victims are blamed for 16 their misfortune (Rubin & Peplau, 1973; R. E. Smith, Keating, Hester, & Mitchell, 1976). Several studies have been conducted in order to investigate the relationship between belief in a just world and victim blaming. Rubin and Peplau (1973) discovered that as belief in a just world increases so does the tendency to assume that the plight of blacks, and of women are deserved. HacLean and Chown (1988) found an association between believing in a just world and blaming elderly people for their poor health and low income. It is important to note that some investigations obtained results that were not consistent with the just world hypothesis. Kerr and Kurtz (1977), using a sample size of 229 subjects failed to find a significant association between just world beliefs and judgments of victim responsibility in a rape case. Ma and Smith (1985), using a large sample size (over 1000) found a nonsignificant correlation of .06 between just world beliefs and authoritarianism. In Figure 1, it was proposed that survivors of abuse would have had considerable prior exposure to the ”unfairness" in their environment. Thus, abused subjects would be less likely to assume that the world is fair. Individuals who believe that their world is unjust, should be less likely to blame abused children than those subscribing to the just world belief system. l7 ngns_gf_ggntzg1L Levenson (1981) asserted that locus of control should be conceptualized as a multidimensional construct. Specifically, Levenson suggested that locus of control may be divided into three categories: Internality, powerful others, and chance. The powerful others construct is considered to be particularly relevant to the present research as parents may be seen as powerful others. It has been suggested that abused persons may be different from others in the extent to which they believe they can control their own reinforcement. Gold (1986) found that women who had been sexually victimized as children were more likely than nonvictims to attribute bad events to internal, stable, global factors; and to attribute good events to external factors. Barahal, Waterman, and Martin (1981) investigated the social cognitive development of physically abused children. They discovered that abused children (ages 6-8) were significantly more likely to attribute external control of events than were nonabused children. Allen and Tarnowski (1989) found similar results on a sample of children ages 7 to 13 years. Barahal et a1. (1981) suggested that maltreated children have little confidence in their power to impact and shape their experiences. Prior victim blame research has been oriented partially toward looking at the relationship between locus of control and attributions of responsibility. Several early studies 18 found that externals victim blame less than internals (e.g., Paulsen, 1979: Phares & Lamiell, 1975: Phares & Wilson, 1972: Sosis, 1974: Thornton, Robbins, & Johnson, 1981: Zuckerman, Gerbasi, & Marion, 1977). Phares and Lamiell (1975) explained that internal persons view themselves as generally responsible in their lives, so too, do they view others as responsible for their own attainment of reinforcement. So sure were Phares and Lamiell of this tendency that they stated "We already know that responsibility attributed to another by internals is greater than it is by externals -at least in accident paradigms" (Phares 8 Lamiell, 1975, p. 24). The picture is not so clear as the earlier theorists suggested. Many of the early studies were methodologically flawed. For example, in the widely cited Sosis (1974) investigation, attribution of responsibility was assessed by means of a single item. No attempt was made to calculate reliability. In addition, a small sample size was used. The problem of a single item measure of attribution of responsibility was repeated in the Zuckerman, Gerbasi, and Marion (1977) study. Other studies were plagued with biased reporting of results. In Alexander's (1980) investigation, externals blamed victims less than internals. However, externals also blamed assailants less than internals. Yet only the former finding was discussed. The latter was ignored. In the current study, the relationship between 19 locus of control and victim blame was assessed using techniques which kept sampling error and measurement error to a minimum. Thus, a large sample size was used (n a 897): and the victim blame scale that was used had 96 items. In Figure 1, it was suggested that adult survivors of abuse, in comparison to controls, would be less likely to view themselves in control of their lives. That is, they would score lower on internal, and higher on powerful others locus of control. Furthermore, those persons scoring lower on internal locus of control would be less likely to behaviorally child blame (since the major difference between internals and externals lies in their sense of being able to affect change in their lives in a quasi-behavioral sense). In addition, those who attribute the major provider of reinforcement in their lives to powerful others would be more likely to see parents as responsible, and hence less justified for child abuse, and they would be less likely to behaviorally child blame. 2g:sgnal_fiin11azity. The defensive-attribution hypothesis, proposed by Shaver (1970), put forth two important motivating factors that influence people when they evaluate victims of misfortune. Individuals have a need to defend against the possibility that random misfortune may happen to themselves (harm-avoidance). In addition, persons are motivated to defend against the possibility that they 20 will be held responsible if they were to end in a similar fate (blame-avoidance). Investigating blame attributions for rape, Thornton (1984) found that in evaluating personally similar victims, behavioral blame was ascribed more than characterological blame while the reverse was true in evaluating personally dissimilar victims. He explained this in terms of harm- avoidance and blame-avoidance motives. However, Thornton did not make the correct comparisons. For example, there is no reason to suppose that in evaluating personally dissimilar victims, subjects should ascribe more chracterological than behavioral blame. What Thornton should have looked at was a question more fundamental to his assertions: that is, are there differences between_snbjggt§ personally similar and personally dissimilar to victims on each of the behavioral and characterological components of victim blame? In the interest of harm-avoidance motives, individuals seeing themselves as personally similar to the victim, compared to those seeing themselves as personally dissimilar, should be more likely to ascribe behavioral victim blame, determining that something the victim did brought about the consequence. Additionally, in the interest of blame-avoidance motives, individuals seeing themselves as personally similar to the victim, compared to those seeing themselves as personally dissimilar, should be 21 less likely to ascribe characterological blame, thereby avoiding damage to self-esteem. In Figure 1, it was suggested that adult survivors of abuse would see themselves as personally similar to victims of abuse. Subjects seeing themselves as personally similar to the victim should demonstrate high levels of behavioral child blame and low levels of characterological child blame, in comparison with those personally dissimilar. Selfzngzggatign. In investigating rape situations, Libow and Doty (1979) demonstrated that the more severely abused victims tended to have the least punitive attitudes toward their assailants. Moreover, victims exhibited considerable self-blame. These findings were explained in terms of "harm avoidance" motives. Blaming the self and reducing the attacker's responsibility allows for a sense of control and a belief in the future avoidability of negative outcome. Turning to child abuse, several investigators have suggested that abused children commonly blame themselves for their punishment (e.g., Green, 1982; Shengold, 1979). Kempe and Kempe (1976) suggested that abused children believe they deserve to be punished. Amsterdam, Brill, Bell, and Edwards (1979) analyzed the responses of adolescents and young adults. They found that the more severely physically abused subjects were as children, the more likely they were to feel that punishment was deserved. One investigation of abusive 22 treatment from the child's perspective (Hey, Moore, McPhee, 8 Trought, 1986) found abused children (ages 5-12 years) to self-blame in all but the most severe cases. However, Herzberger, Potts, and Dillon (1981) failed to find differences between abused and nonabused children (8-14 years) in terms of self-blame. Several studies suggest that the abused child typically has a low self-image and regards him/herself with self- deprecation and contempt (Green, 1982). Hjorth & Ostrov (1982) found that abused adolescents (ages 12-16 years) had poorer overall self-image than did matched controls. Allen and Tarnowski (1989) found that physically abused children (ages 7-13 years), compared to matched controls, demonstrated lower levels of self-esteem. Cohen (1984) suggested that abused children may blame themselves for the abusive experience and may self-derogate in order to justify their parents' behaviors. Cohen went on to say that the abused child may see him/herself as "bad" rather than seeing the parent as inappropriate. Helfer (1987) suggested that within the family unit, abused children often are forced to accept responsibility for the behaviors and inadequacies of their parents. This pattern is referred to as role- reversal. These children are expected to nurture/comfort their parents and to take on many aspects of the parenting role. 23 In Figure 1, it was suggested that adult survivors of abuse would indicate that when they were children (under 17 years of age), they derogated themselves (i.e., they saw themselves as ”bad” children). Part of self-derogation consists of self-blame for parental punishment. It was suggested that this view of having been a bad child should persist into adulthood. As such, individuals indicating childhood self-derogation and self-blame should presently see themselves as having been bad children and would self- blame for the parental punishment they had received as children. Such individuals should be likely to consider other children as blameworthy for parental punishment. Hence they would be more likely to child blame. The processes indicated in Figure 1 were expressed in terms of specific hypotheses presented below. W 7. Adult survivors of physical abuse, in comparison to individuals having experienced less abuse, should demonstrate less cognitive empathy, and less emotional empathy. 8. Individuals scoring lower on cognitive and emotional facets of empathy should score higher on child blame. 9. Adult survivors of physical abuse, in comparison to individuals having experienced less abuse, should demonstrate lower levels of just world beliefs. 24 10. Individuals indicating lower levels of just world beliefs, should score lower on child blame. 11. Adult survivors of physical abuse, in comparison to individuals having experienced less abuse, should demonstrate lower levels of internal locus of control, and higher levels of powerful others locus of control. 12. Individuals indicating lower levels of internal locus of control should score lower on behavioral child blame. 13. Individuals demonstrating higher levels of powerful others locus of control should score lower on abuser justification and lower on behavioral child blame. 14. Adult survivors of physical abuse, to a greater extent than individuals having experienced less abuse, should see themselves as personally similar to abused children. 15. Individuals indicating higher perceptions of personal similarity to abused children should score higher on behavioral child blame, and lower on characterological child blame. 16. Adult survivors of physical abuse, in comparison to individuals having experienced less abuse, should indicate having experienced during their childhoods, more self-derogation and self-blame for parental punishment. 17. Individuals indicating more childhood experience of self-derogation and self-blame for parental punishment 25 should demonstrate more current derogation of their childhood selves and self-blame for parental punishment. 18. Individuals indicating more current derogation of their childhood selves and self-blame for parental punishment should demonstrate more child blame. The current study was an investigation of the victim blame process in adult survivors of child physical abuse. Several factors were proposed as intervening variables in this process. Based on the prior literature, it was suggested that some of the intervening variables would act as paths by which child physical abuse leads to increases in child blame (e.g., empathy, self-derogation, self-blame: see Figure 1). However, some of the intervening variables were proposed to act as paths by which child physical abuse leads to decreases in child blame (e.g., just world beliefs, locus of control). As such, at the onset of the current study, the proposed path model suggested an;aggnig;ig_gff§gt§ (Hunter, 1986). Consequently, there was no prediction made as to whether child physical abuse, overall, leads to increases or decreases in child blame. No prior research has investigated this question directly. In a related area, child sexual abuse, Waterman and Foss-Goodman (1984) found that adult survivors of abuse were less likely to victim blame than controls. However, no theoretical explanation was given for this finding. 26 WW According to Gelles (1976) as well as Pelton (1978), contrary to some pronouncements (cf. Steele, 1975), child abuse is not evenly distributed across the social structure, but rather is strongly related to poverty, in terms of prevalence and of severity of consequence. It appears that abuse occurs more frequently among lower SES families (Trickett, Aber, Carlson, & Cicchetti, 1991). Simons, Whitbeck, Conger, and Chyi-In (1991) found that parents of low education are more likely to utilize harsh discipline in raising boys. Toro (1982) asserted that it is necessary for any study looking at the consequences of childhood abusive experience to consider the effects of $88 as a possible confounding factor. As such, the current study investigates the path from childhood experience of abuse to victim blame, with the effects of SBS held constant. 27 Additional Issues Explored MW Other than childhood experience with abuse, another important factor that may lead subjects to see themselves as personally similar to victims of abuse is "gender concordance.” That is, individuals judging victims of their own gender should be more likely to see such victims as personally similar. Skrypnek (1980) found that in judging female rape victims, females ascribed more blame of the behavioral type than did males. However, it is not clear from the Skrypnek (1980) study whether the variable of influence is gender per se (i.e., females simply ascribe more blame), or alternatively gender concordance (i.e., females judging females feel a sense of personal similarity to the victim, and in the interest of harm avoidance motives behaviorally blame them). It appears that the variable operating in Skrypnek’s investigation is that of gender concordance as much evidence has accumulated indicating that in general, males tend to victim blame more than females (e.g., Calhoun, Selby, Cann, 8 Keller, 1978; Deitz, Blackwell, Daley, 8 Bentley, 1982; Herzberger 8 Tennen, 1985; Kanekar 8 Kolsawalla, 1980; Waterman 8 Foss-Goodman, 1984). In the current study, it was suggested that individuals judging victims of their own gender would experience a sense of personal similarity to the victim. 28 Hypothg§i§_12‘, Individuals judging victims of their own gender (gender concordant), in comparison to those judging victims of the opposite gender (gender discordant) should indicate higher perceptions of personal similarity to abused children. METHOD thiflfitfi Subjects consisted of 897 undergraduate students at Michigan State University recruited via the human subject pool. Students participate in the human subject pool as part of either course requirement or to earn extra credit (introductory psychology). No monetary inducements were offered to participants. Sixty-eight percent of the subjects were female. Most (87.6 %) of the subjects were Caucasian. Blacks were represented by 7.0 t of the subjects. Of the total sample, 866 individuals did not have any children. Analyses were restricted to these persons. The mean age of these students was 18.9. On the Duncan Socio-economic Index (Hauser 8 Featherman, 1977), the mean and standard deviation socio-economic status (SES) scores were 57.47 and 19.24 respectively. The current sample indicated SES levels fairly close to those of the normative sample. By means of the Conflict Tactics Scales (CTS: Straus, 1989), the 866 participants were placed on a continuum from non-abused to extremely abused. It was expected that on the ”severe violence” index of the CTS, at least 170 of the subjects would endorse having experienced such parental punitiveness at least once. Furthermore, it was expected 29 30 that based on the overall rate of child abuse in the general population (Straus, 1989: Straus 8 Gelles, 1986) and based on reported histories of college students (Berger, Knutson, Mehm, 8 Perkins, 1988: Miller 8 Miller, 1983), approximately 3-98, or 26-78 participants out of the 866 subjects, would indicate having experienced "very severe violence" on the Conflict Tactics Scales. Of the 866 students, 323 individuals endorsed having experienced at least one incident of "severe violence" (on the Conflict Tactics Scales) from at least one parent. In addition, the number of subjects who endorsed having experienced "very severe violence" corresponded closely to the findings reported by Straus (1989). For example, the number of subjects who endorsed having been beaten up at least once by their fathers and mothers respectively were n - 22, and n - 14. The number of subjects who endorsed having had a knife or gun used on them at least once by their fathers and mothers respectively were n - 21, and n = 9. Materials An adapted version of the WW8). (Straus, 1989: Straus 8 Gelles, 1989) was used in order to provide an indication of the respondent's childhood experience with physically abusive parenting. The measure listed 25 possible ways to deal with conflict ranging from discussing the issue calmly to using a knife or gun (see 31 Appendix A). Subjects indicated how frequently each conflict tactic was used during their childhood (under 17 years of age). Conflict tactics were reported separately for each parent. Although the CTS has not been used typically for the purposes of assessing childhood experience of abuse, the recent research of Berger and colleagues (Berger, Knutson, Mehm, and Perkins, 1988) suggested that if parental punitiveness is broken down is terms of specific behaviors, subjects are able to provide self-reports that are reliable and valid measures of prior experience of child abuse. The CTS has 3 subscales. These are minor violence, severe violence, and psychological aggression. For the purposes of the current study, a neglect subscale was developed. Coefficient alpha reliabilities for the severe violence index (items 20-25; see Appendix A), have been demonstrated to be .49 (Straus, 1989). Both the severe violence index and the gang severe violence index (items 20, 8 22-25) are considered to be measures of child abuse (Straus, 1989). For the psychological aggression scale (items 7, 8, 9, 11, 15, 8 16) coefficient alpha reliabilities have ranged from .62 to .77. The construct validity of the CTS has been reasonably well documented. For example, there is broad consensus that stress increases the risk of child abuse. Research using the CTS has yielded results consistent with that theory (Straus, 1989). 32 Several items (12, 13, 8 14) were added to the psychological aggression subscale. These items are: 1) Did or said something to hurt your feelings. 2) Did or said something to embarrass or humiliate you. 3) Did or said something psychologically cruel to you. Items 4, 5, and 6 form the neglect subscale that was developed for the current study. These items are: 1) Ignored you for a day or two. 2) Ignored you for several weeks at a time. 3) Would not make food or shelter or clothing available to you. The W (Davis. 1980. 1983) is a 28-item, self-report measure consisting of four 7-item subscales, each tapping some aspect of the global concept of empathy. Subjects respond to self-descriptive statements on a 5-point intensity scale ranging from ”does not describe me well" to "describes me very well". This questionnaire was built on the assumption that empathy is a multidimensional construct. The four subscales consist of: (a) perspective-taking (PT), (b) fantasy (FS), (c) empathic concern (EC), and (d) personal distress (PD). The PT scale assesses the tendency to adopt spontaneously the psychological point of view of others: the FS scale taps respondents' tendencies to transpose themselves imaginatively into the feelings and actions of fictitious 33 characters in books, movies, and plays. The EC scale assesses "other-oriented" feelings of sympathy and concern for unfortunate others. The PD scale measures "self- oriented” feelings of personal anxiety and unease in tense interpersonal settings. Internal reliabilities for these subscales range from .71 to .77, while test-retest reliabilities range from .62 to .71 (Davis, 1980). Davis (1983) found the PT scale to be positively correlated (r = .40) with the Hogan Empathy Scale (Hogan, 1969), a cognitiye empathy measure. Davis (1983) also found the FS and EC scales to be positively correlated (mean rs of .52 and .60 respectively) with the Mehrabian and Epstein Empathy Scale (Mehrabian 8 Epstein, 1972), an emotional empathy measure. The in§;_flg;1d_§gnle adapted from Rubin and Peplau (1973) is a 16-item questionnaire designed to measure the extent of an individual’s belief that the world is fair. Subjects indicate the extent of their agreement with each statement by means of a 7-point intensity scale ranging from strongly disagree to strongly agree. Rubin and Peplau (1973) calculated a coefficient alpha reliability estimate of .79, for a college student sample. O'Quin and Vogler (1990) found coefficient alphas of .70 and .38 for college students and prison inmates respectively. The 1,_2,_and_g_sgalgs (Levenson, 1981) were built on the assumption that locus of control is a multidimensional construct. The questionnaire measures the extent to which 34 the subject believes: a) s/he has control over his/her own life (I Scale), b) powerful others have control over his/her life (P Scale), c) chance forces have control over his/her life (C Scale). Subjects indicate on a 6-point intensity scale the extent to which they agree with each of the 24 statements (8 statements for each of the three scales) regarding the locus of control in their lives. In one investigation, Kuder-Richardson reliabilities yielded .64, .77 and .78 for I, P, and C Scales respectively. Spearman- Brown split half reliabilities have been found to be .62, .66 and .64 for I, P, and C scales respectively. The validity of the I, P, and C Scales has been demonstrated through convergent methods. Rotter’s I-E scale correlates positively (.25, .56) with both the P and C scales respectively and negatively (-.41) with the I scale (Levenson, 1981). The sg1fgngzggatign_1ndgx_(snll is a 12-item inventory developed for the purposes of the present investigation. Responses are made on a 5-point intensity scale ranging from ”not at all" to "considerably". Six of the items assess the extent to which, during their childhoods, subjects experienced self-derogation and self-blame for parental punishment. These items are: 1) When you were a child to what extent did you see yourself as a bad child? 35 2) When you were a child, to what extent did you see yourself as a naughty child? 3) When you were a child, to what extent did you see yourself as a disobedient child? 4) When your parents punished you as a child, to what extent did you blame yourself for what you got? 5) When your parents punished you as a child, to what extent did you feel you got what you deserved? 6) When your parents punished you as a child, to what extent did you feel responsible for what you got? Six additional items assess the extent to which subjects experience current derogation of their childhood selves and current self-blame for former parental punishment. These items are: 1) To what extent do you now see yourself as having been a bad child? 2) To what extent do you now see yourself as having been a naughty child? 3) To what extent do you now see yourself as having been a disobedient child? 4) Thinking back to when your parents punished you, to what extent do you now blame yourself for what you got? 5) Thinking back to when your parents punished you, to what extent do you now feel you got what you deserved? 6) Thinking back to when your parents punished you, to what extent do you now feel responsible for what you got? 36 (see Appendix B). The W is a multidimensional measure of blame developed by the author for this study. In accordance with Janoff-Bulman's (1979) multidimensional conceptualization of blame, the construct of victim blame is measured in terms of characterological and behavioral components. The questionnaire consists of 8 scenarios of child abusive situations, parts of which were derived from Stollak, Scholom, Kallman, and Saturansky (1973). These vignettes vary based on the level of provocativeness (high vs. low) of the child in the story, gender of the child, and gender of the abuser. Highly provocative children are verbally aggressive and insolent. Very mildly provocative children perform minor I transgressions such as getting clothing dirty. In each situation, the parent's behavior is seen as abusive in accordance with Gil's (1970) definition of child abuse (the use of physical force in training and punishing children). The definition of child abuse used for these scenarios is also consistent with Helfer's (1987) definition (any interaction or lack of interaction between family members which results in some non-accidental harm to the individual’s physical and/or developmental states). Responses for each of the items are made on a 7-point Likert scale. Following each of the 8 scenarios, subjects indicate (on a single item) the extent to which they 37 perceive the child in the vignette to be provocative (a manipulation check). Following this, subjects state (on a single item) the extent to which they perceived the event to be severe and traumatic for the child (it is assumed that each of these events is highly severe and traumatic). Next, subjects indicate the extent to which they agree/disagree with four statements designed to measure behavioral child blame. These statements are: 1) X was responsible for what happened to him/her because of what s/he did or said. 2) X got what s/he deserved because of what s/he did or said. 3) The events of that day were X’s fault because of what s/he did or said. 4) X is to blame for what happened to him/her because of what s/he did or said. Subjects then indicate the extent to which they agree/disagree with four statements designed to measure characterological child blame: 1) X was responsible for what happened to him/her because of the kind of person s/he is. 2) X got what s/he deserved because of the kind of person s/he is. 3) The events of that day were X's fault because of the kind of person s/he is. 38 4) X is to blame for what happened to him/her because of the kind of person s/he is. Following this, subjects state the extent to which they agree/disagree with four statements measuring abuser justification: 1) Y’s behavioral response toward X was justified. 2) Y's actions toward X were appropriate to what X did. 3) Y really had no alternatives other than what s/he did. 4) Most people would have responded as Y did. Finally, subjects indicate the extent of their agreement or disagreement with four statements measuring their sense of personal similarity to the victim. These statements are: l) I feel similar to X. 2) I identify with X. 3) X reminds me of myself. 4) I feel a sense of sameness with x. For the 16 aforementioned items, "X" and "Y" was be replaced respectively by the names of the child and parent in the vignette (see Appendix C). The W (Ward. 1988) is a 25 item questionnaire which is designed to tap supportive/favorable and unsupportive/unfavorable predispositions toward rape victims with special emphasis on 39 those attitudes which reflect disbelief, blame, or denigration, and/or trivialization of the seriousness of rape and its effects on victims. Subjects indicate on a 5- point intensity scale the extent of their agreement with specific attitudinal statements with respect to rape. The ARVS has been found to have a Cronbach's alpha reliability of .83 to .86, and test-retest reliability of .80 (Ward, 1988).. The ARVS's construct validity was demonstrated through correlations of the ARVS with Burt's (1980) Adversarial Sexual Beliefs scale (I = .41, p < .05) as well as Burt’s Acceptance of Interpersonal Violence scale (I = .26, p < .05). . The NEW (see Hauser 8 Featherman, 1977) was used to measure socio- economic status. It is the most commonly used single SES measure (Campbell 8 Parker, 1983). It obtains three pieces of information: occupation, industry, and class of worker. Subjects were asked to provide this information with respect to the occupations of their fathers and mothers separately. These data were subsequently coded into the Duncan Socioeconomic Index using the coding system found in Hauser and Featherman (1977). Using this procedure, it was possible to assign explicit SEI scores according to standardized conventions, eliminating most subjective coding judgments. Following questions pertaining to SES, subjects responded to several basic demographic questions, including 4O gender, age, race, religion, whether the subject has children of his/her own (and if so, how many), the number of male and female siblings, and whether the subject has been in psychotherapy (including the number of sessions) (see Appendix D). RIQQEQHIB Prior to the administration of questionnaires, interrater reliability on the provocativeness of the scenarios was established using 9 volunteer colleagues (graduate students) who were blind to the hypotheses. Students participating in research through the human subject pool had the option to participate in the current investigation. Several group administrations were conducted. Subjects choosing to attend only one administration were provided with all the aforementioned questionnaires to complete at that time. Subjects electing to attend a second administration were provided with all the questionnaires to complete on the first occasion; they completed only the Child Abuse Scenarios Questionnaire and the Self-Derogation Index on the second (for the purpose of measuring test-retest reliability). Each set of questionnaires administered on the first occasion had a specific ID number associated with it. The ID number was written in large letters on the cover page of the set. Subjects were instructed to tear off, and keep the cover page. They were informed that those wishing to return 41 for the second administration must bring their ID number cover pages with them, so that their pair of questionnaires could be appropriately matched. The 8 vignettes from the Child Abuse Scenarios Questionnaire were randomly ordered. On the first administration, a letter of consent (see Appendix E) was given first, followed by the Child Abuse Scenarios Questionnaire, the I, P, and C Scales, the Just World Scale, the Conflict Tactics Scales (father then mother), the Interpersonal Reactivity Index, the Self-Derogation Index, the Duncan Socio-economic Index, and the Attitudes Toward Rape Victims Scale. On the second administration, a page of instructions (Appendix F) was administered first, followed by the Child Abuse Scenarios Questionnaire and the Self- Derogation Index. One month following the final data administration, a one page explanation of the main ideas behind this study (see Appendix G) was available at a central office in the department of psychology. At the time of their participation, all subjects had been notified of the date that this handout would become available. Once all data was collected, the coding of the Duncan Socio-economic Index commenced. Three research assistants received training in the coding system for the Duncan SEI (in Hauser 8 Featherman, 1977) until they achieved proficiency. Training consisted of bi-weekly sessions for 42 approximately 1 month, in which sample protocols were discussed and problems were clarified. Upon completion, trainees achieved an interrater reliability level of r (Spearman-Brown) - .78, p,< .001. Following the collection of all data, the three assistants divided the 897 questionnaires for coding of the Duncan SEI. RESULTS AND DISCUSSION The data were analyzed and evaluated in several stages. First, a manipulation check was conducted on the dimension of provocativeness of the vignettes. This manipulation check was conducted as a pre-test on 9 volunteer graduate students. Second, the contribution of several factors to increases in child blame (study I: hypotheses 1-4) were analyzed using the Analysis of Variance (ANOVA) technique. Third, the measurement model was analyzed using confirmatory factor analyses. In doing so, the construct dimensionality of victim blame (study II: hypotheses 5-6) was examined. Fourth, the path from childhood experience of abuse to child blame (study III: hypotheses 7-18) was examined using correlation coefficients and the method of path analysis. Fifth, the effect of gender concordance on personal similarity (hypothesis 19) was also examined using ANOVA. All hypothesis testing was conducted on the sub-sample of 866 undergraduates at Michigan State. E !' H i J l' m l Interrater reliability on the provocativeness of the scenarios was established using 9 volunteer colleagues (graduate students) who were blind to the hypotheses. These judges were administered the 8 vignettes from the Child Abuse Scenarios Questionnaire in random order, and were 43 44 asked to indicate the extent to which they viewed the child in each story as provocative. The judges' ratings were correlated with one another: interrater reliability was calculated to be): (Spearman—Brown) - .95, p < .001. On a 7-point intensity scale ranging from "strongly disagree" (1) to "strongly agree" (7), judges indicated (for each scenario) the extent to which they agreed with the assertion that the child’s behavior was provocative. For the 4 scenarios assumed a prigni to represent provocative children, judges ranged in their rating from 5.11 (agree somewhat) to 5.77 (agree). For the 4 scenarios assumed a priori to represent non-provocative children, judges ranged in their rating from 2.55 (disagree somewhat) to 1.77 (disagree). Several analyses of variance indicated that the 4 provocative scenarios were rated as significantly more provocative than the 4 non-provocative scenarios (F (1, 56) - 147.38, p < .001). In addition, the least provocative of the 4 provocative scenarios was judged to be significantly more provocative than the most provocative of the 4 non- provocative scenarios (F (1, 56) = 22.78, p < .001). In the sections to follow, the results from each study are presented and discussed. An overall discussion will follow the study by study presentation. 45 Study I: Factors Contributing to Child Blame Results Four factors were tested for their influence in the manifestation of child blame. Three of these factors were within-groups experimental variables that were manipulated as follows. Eight vignettes of child abusive situations were given to each subject. The scenarios differed on the three completely crossed variables: provocativeness of the child in the scenarios, gender of the abuser, and gender of the child. The fourth factor tested was a between-groups individual difference variable, gender of the subject. The influence of these four variables was tested statistically by means of a mixed-model Analysis of Variance (ANOVA) procedure, with the independent variables being gender of the subject, provocativeness of the child (high vs. low), gender of the abuser, and gender of the child. The dependent variable was child blame. The results of this AHOVA are presented in Table 1. The means and standard deviations for the ANOVA are presented in Tables 2 and 3. 46 Table 1 Analysis of Variance of Child Blame: Subject Gender by Child Provocativeness by Abuser Gender by Child Gender 3992:: SS DE HS 2 Eta A (Subject Gender) 211.4 1 211.4 40.1** .15 S/A (Subj. Within 4316.7 819 5.3 .69 Gender) B (Provocativeness) 1472.9 1 1472.9 1198.6** .40 B x A 12.2 1 12.2 9.9* .04 B x S/A 1006.5 819 1.2 .33 C (Abuser Gender) 91.8 1 91.8 238.2** .10 C x A .5 1 .5 1.2 .01 C x S/A 315.6 819 .4 .19 D (Child Gender) 62.2 1 62.2 l98.7** .08 D x A 1.6 1 1.6 5.1* .01 D x S/A 256.5 819 .3 .17 B x C 43.4 1 43.4 128.0** .07 B x C x A 3.0 l 3.0 8.8* .02 B x C x S/A 277.4 819 .3 .18 B x D 71.4 1 71.4 216.9** .09 B x D X A .0 1 .0 .0 .00 s x D x S/A 269.6 819 .3 .17 C x D 102.9 1 102.9 326.9** .11 C x D x A .0 1 .0 .1 .00 c x D x S/A . 257.9 819 .3 .17 B x C x D .3 1 .3 .8 .01 B X C X D x A .6 1 .6 2.0 .01 B X C x D x S/A 247.6 819 .3 .17 * p < .05 .* n < .001 47 Table 2 Group Means and Standard Deviations of Child Blame Broken Down by Provocativeness, Abuser Gender and Child Gender (Female Subjects only) Abuser Gender Female Male Abuser Abuser M SD M SD Nonprovocative Child Female Child 1.75 .82 1.53 .69 Male Child 1.47 .62 1.72 .76 Provocative Child Female Child 2.28 1.12 2.44 1.18 Male Child 2.38 1.19 3.15 1.40 N.B. n = 579 Table 3 Group Means and Standard Deviations of Child Blame Broken Down by Provocativeness, Abuser Gender and Child Gender (Male Subjects only) Abuser Gender Female Male Abuser Abuser M SD 3 SD Honprovocative Child Female Child 1.98 .81 1.84 .86 Male Child 1.71 .76 2.14 .96 Provocative Child Female Child 2.74 1.22 2.88 1.21 Male Child 2.95 1.28 3.63 1.40 N.B. n,- 242 48 The results presented in Table l were used to test hypotheses 1-4 as follows. Hypothesis_1 stated that male subjects, in comparison to females, should ascribe more child blame. The results demonstrated that males, compared to females, indicated significantly higher (F (1, 819) - 40.1, p < .001, eta - .15) child blame. Mean child blame ratings for male and female subjects respectively were 2.48 and 2.09. Hypgtne§i§_2 asserted that when rating provocative, in comparison to nonprovocative children, subjects should ascribe greater child blame. The results indicated that provocative children were rated significantly higher (F (1, 819) = 1198.6, 9 < .001, eta a .40) than were nonprovocative children. Mean child blame ratings for provocative and nonprovocative children respectively were 2.70 and 1.70. H!Dfl§h§§i§_3 proposed that when rating male, in comparison to female abusers, subjects should indicate more abuser justification. Since the three child blame scales were found to be specific factors of an overall child blame dimension, this hypothesis was adjusted to reflect that finding. The results demonstrated that when rating stories of male, in comparison to female abusers, subjects indicated significantly more (F (1, 819) = 238.2, p < .001, eta - .10) overall child blame. Mean child blame ratings for stories with male and female abusers respectively were 2.33 and 2.08. 49 Hypothesi§_4 stated that when rating male, in comparison to female children, subjects should indicate higher levels of child blame. The results demonstrated that male children were rated significantly higher (F (1, 819) a 198.7, p,< .001, eta - .08) than were female children. Mean child blame ratings for male and female children respectively were 2.30 and 2.10. Ireatment_lnteractign_fiffests Although the Analysis of Variance revealed the four main effects that were hypothesized a priori, it also suggested several possible interaction effects. The 3 largest ones were provocativeness by abuser gender, provocativeness by child gender, and abuser gender by child gender. These were analyzed further using planned comparison ANOVAs. W. The results suggested that for scenarios with nonprovocative children, stories with male abusers, in comparison to female abusers, were rated significantly higher on child blame (F (1, 859) = 8.4, p < .004). However, the effect size was quite small (eta - .06). Similarly, for vignettes with provocative children, stories with male abusers, in comparison to female abusers, were rated significantly higher on child blame (F (1, 855) - 374.6, 9 < .001). However, the effect size was relatively large (eta a .34). 50 2zgygggtiygng§§_py_gn11g_§gndez. Results indicated that for stories with nonprovocative children, there was no significant difference on child blame ratings for male and female children. However, for scenarios with provocative children, child blame ratings for male children were significantly higher than those for female children (F (1, 855) - 333.1, p < .001, eta - .33). Abnsgr_§endez_hy;§h11g_§gnggz. The results suggested that for scenarios with female children, there was no significant difference on child blame ratings for vignettes with male and female abusers. However, for stories with male children, child blame ratings for scenarios with male abusers were significantly higher than for vignettes with female abusers (F (1, 854) a 567.8, p < .001, eta c .28). Trsatment_bx_Subissts_lnterastion_8ffssts The results of the Analysis of Variance suggested several treatment by subjects interaction effects on child blame. The largest of these, provocativeness by subjects, had a relatively large effect size (eta) of .33. An imperfect correlation (r a .75, corrected for attenuation) between child blame scores for provocative and nonprovocative scenarios further suggested a provocativeness by subjects interaction. It was posited that these results may imply that provocativeness is interacting with some outside variable in the elicitation of child blame. The relevance of this implication is that if indeed such an 51 interaction is taking place, then any further individual difference analysis using the child blame dimensions, must be conducted separately for provocative and nonprovocative scenarios. Further analyses revealed two important pieces of information. First, the variance on child blame scores was 50% higher for scenarios with provocative children, than for those with nonprovocative children. This suggested a broader range of subject response for scenarios with provocative children. Second, a scatterplot (see Figure 2) of the relationship between child blame scores for provocative and nonprovocative children, demonstrated an increasing nonlinear function, which suggested floor effects on child blame scores for nonprovocative scenarios. Thus, the provocativeness by subjects interaction effect appeared to be an artifact of the child blame scale, rather than the result of an unexplained variable. Consequently, subsequent analyses using the child blame dimension did not need to be conducted separately for provocative and nonprovocative scenarios. 51A magma «.5339. TB 935 now $5 soggaehsflflmshfiauflusm 888% ABE gmgv 85m 36 I.6 H.‘ n.u 0.. l.n IoN ..H d. 0.I + H H H HH H HNQOHNHOOOOQNOOIHI +H H H H N HHQNNHHNHNONNOON H H HNHHH N HvNHNQnHanannNNNN H H H H HH H NnNNnd OHNHVHHNONHN H H H H H H NHHH HHH HHNH nHNnHN H H 4 H HH HH 6 HH 6 HvONvNNN HHHN +n.H H HHHH HHHHHHH H NNN NH H H H H H HH HH HNO 6 HHNHN HNn nnnN'N n H H H H N n H CHHHNNQHHNQNNHHNHGNHHHH H H H H NH NH HH NH GH nN 0 HH HNH H 4 HH HH N HH nv'HHH anH N nNnH H 4N H HNHH NN HNH H HNHQ H H H H H H HH H N H HN NH H. H N H H H H NvH NN H HH NHHHHHN N H H HH H HHN NH N NV 0 HH N H + H N H N H NH NHHH H H +D.N H HH HH H HN HN H H H H H H H H HNHHHHNNH HH H H H H H HH NN N0 H H H H H H NH H N H H H + H H H N H H N H H +0 H N HHH HH H H H N HNH H H H H H H H H H H H H H H + N H H +0.0 H H HH H H H H H H H H N H H 9 +1 H H H H H H H H H H 4 +D.Q H H H H H H H H H 9 OD bluuue ASE gfigv 85m "55 D I On I I ON 4 I O F I ON 6 I OH O I O I I ON H I OH O I I N I 6N I I 6H 9 I b I In 0 I IN 9 I OH 0 I O I ‘H N I ON h I OH I I O I on O I vN I I 0H 0 I v I an I I RN 0 I HH 0 I H I NH I I NN 0 I NH N I N I HH d I HN I I HH H I H H 52 Study I: Factors Contributing to Child Blame Discussion W. It was suggested that male subjects would attribute greater victim blame than would female participants. The results corroborated that assertion. Earlier researchers investigating victim blame demonstrated that males were more likely to blame rape victims (Calhoun et al., 1978; Deitz et al., 1982). Investigating attributions of responsibility for situations of harsh parental discipline, Herzberger and Tennen (1985) discovered that females were more likely to view harsh parental discipline as inappropriate. As such, the current results corroborated the findings of earlier research. Wm. This hypothesis asserted that provocative children would elicit higher levels of child blame. The results indicated large differences (in the hypothesized direction) in child blame ratings given to provocative and nonprovocative children. These findings are particularly interesting in light of earlier research in this area. In an investigation of 570 German families, Engfer and Schneewind (1981) found that having a child that is perceived as difficult to handle (i.e., "problem child") is the best predictor of harsh parental punishment. The research of Herrenkohl et a1. (1983) pointed to similar conclusions. Reid et al. (1981) 53 reported that hyperactive children are more likely to be physically abused. The results of the current investigation and the findings of earlier studies point to the conclusion that provocative children are more likely to receive abusive parenting: and they are more likely to be blamed for these parental behaviors. Two inferences may be drawn from this assertion. First, it is possible that cognitive perceptions and attributions play a role in the manifestation of physical child abuse. Parents may justify their abusive behaviors by viewing the provocative child as "deserving" of harsh punishment. Larrance and Twentyman (1983) and Herrenkohl et a1. (1983) stressed the importance of cognitive perceptions as an antecedent to physical child abuse. A second, more speculative inference may be drawn from the current findings. This study demonstrated that provocative children received more child blame than nonprovocative children. Consequently, abusive parents of provocative children may be more likely to blame their children than abusive parents of nonprovocative children. If that is so, and if these parents communicate their attributions of responsibility to their children, then it may be the case that provocative children, to a greater extent than nonprovocative children learn that they were responsible for their parents' actions. Children who truly 54 were provocative at a young age may be more likely to grow up to believe that they were responsible for the harsh punitive responses of their parents. W. It was suggested that male abusers would elicit greater sentiments of abuser justification. The results supported this hypothesis. When responding to child abusive scenarios with male, in comparison to female parents, subjects ascribed higher levels of child blame (which included abuser justification). These findings corroborate those of earlier studies (Herzberger 8 Tennen, 1985) that found similar main effects for abuser gender. Herzberger and Tennen discussed their results in terms of the stereotyped expectation that mothers should be nurturant and loving. They stated that the belief that severe maternal treatment is unexpected and unique may lead observers to view such discipline as particularly inappropriate and undesirable. In addition to the main effect of abuser gender discussed above, the results revealed several interesting interaction effects. That is, subjects ascribed greater child blame when reading scenarios with male, in comparison to female abusers. However, this effect was much stronger when the child was provocative. This result may be explained as an artifact of the floor effect on child blame scores for nonprovocative scenarios. When reading scenarios of nonprovocative children, child blame scores were all 55 degraded. Thus differences in abuser gender may have been masked. A second interaction effect with respect to abuser gender was as follows. Male abusers elicited greater sentiments of child blame. However, this result was particularly pronounced for male children. These results are not consistent with those of Herzberger and Tennen (1985) who found no such interaction effects. One of the primary differences between the current investigation and that of Herzberger et al. was the sample size. Herzberger et a1. had 86 participants. With such a sample size, the interaction effect currently being discussed would not have been detectable. The results of the current study imply that male children who are abused by male parents are at considerable risk for child blame. In court cases involving such combinations of parent and child, it may be necessary to pay particularly close attention to the possibility of judicial bias. WWW- This hypothesis stated that when rating male, in comparison to female children, subjects should indicate higher levels of child blame. The results supported this assertion. The findings are consistent with those of Waterman and Foss-Goodman (1984). In studying child sexual abuse, they found a tendency for male children to be blamed more than female 56 children. The prevailing cultural myth of girls being ”sweet and innocent" may underlie this tendency. Prior research has demonstrated that boys are more likely candidates for physical abuse than girls (Gil, 1970: Lieh-Mak, Chung, 8 Liu, 1983: Maden 8 Wrench, 1977). The results of the current investigation taken together with the findings of earlier studies suggest that boys receive more abusive parenting: and they are more likely to be blamed for these parental behaviors. As with provocativeness, two inferences may be drawn from these notions. First, it is possible that, once again, cognitive perceptions and attributions play a role in the manifestation of physical child abuse. Parents may rationalize their harsh punitive behaviors by viewing the male child as "deserving" of punishment. A second inference that may be drawn from the current findings is as follows. This study demonstrated that male children received more child blame than female children. Consequently, abusive parents of male children may be more likely to blame their children than abusive parents of female children. If these parents communicate their attributions of responsibility to their children, then it may be the case that boys, more than girls come to believe that they were responsible for their parents' behaviors. Among abused children, boys should be more likely to blame themselves for parental punishment. 57 In addition to the main effect of child gender noted above, the findings demonstrated one interaction effect. Specifically, participants indicated more child blame to boys than to girls. But this effect was existed only when the child was provocative. Once again, this result may be explained as an artifact of the floor effect on child blame scores for nonprovocative stories. In rating vignettes of nonprovocative children, child blame levels in general were lowered. As such, differences in child gender may have been covered up. 58 Study II: The Construct Dimensionality of Victim Blame Results Several earlier researchers (Hunter 8 Gerbing, 1982: Hunter, Gerbing, 8 Boster, 1982) have suggested that in order to conclude that a scale is structurally unidimensional, the correlations among the items should conform to the covariance structure (Joreskog, 1978) of a unidimensional measurement model as evaluated by a confirmatory factor analysis. In such a case, the correlation matrix should be consistent with two product rules (Hunter 8 Gerbing, 1982): internal consistency and external consistency (parallelism). If the correlations among the items within a cluster form a Spearman Rank 1 matrix, they are said to be internally consistent (Green, Lissitz, 8 Mulaik, 1977). Items are externally consistent if their correlations with all variables outside the cluster are directly proportional to one another (Hunter, Gerbing, 8 Boster, 1982). In order to analyze the measurement model of the current investigation, one estimation procedure was used. A multiple-groups centroid factor analysis (confirmatory factor analysis) was calculated using PACKAGE (Hunter 8 Cohen, 1969: Hunter, Gerbing, Cohen, 8 Nicol, 1980). Communalities were computed by iteration within each cluster. The cluster solution was sought by successively 59 repartitioning the items until the criteria of unidimensionality was achieved for each cluster. In the following section, the results from the confirmatory factor analyses will be presented. For the majority of the scales, the presentation of these results will be very brief. However, since one of the purposes of the current thesis was to investigate the construct dimensionality of victim blame, the results from the Child Abuse Scenarios Questionnaire (CASQ) and the Attitude Toward Rape Victims Scale (ARVS) will be considered in greater detail. Finally, the scale reliabilities, means, and standard deviations will be presented for all measures. QQniiImaIQIx1EBQIQI.AD§1¥E§S.B§§HLLS QQnf1ict_1actigs_§ca1gs_L§Ifil. Subjects reported their fathers' and mothers' conflict tactics separately. Each of the items were initially dichotomized into 0 (never happened) versus 1 (happened at least once). For each parent separately, four CTS subscales were examined. These were minor violence, severe violence, psychological aggression, and neglect. For all 8 subscales, no items were rejected on the basis of either internal or external consistency. Since the CTS was derived as a Guttman scale (Straus, 1989), it was expected that each subscale would demonstrate cluster properties consistent with the Guttman scale model. The findings corroborated that expectation. For reports of both paternal as well as maternal conflict 60 tactics, the minor violence and severe violence subscales demonstrated parallelism to one another with respect to all outside variables. This implies that a single physical abuse dimension underlies these two scales. In addition, when a frequency count was taken for each item, the results demonstrated no clear division in the distribution of minor violence and severe violence items. In other words, minor and severe violence items were distributed as a single continuous (yet non-normal) scale. As such, they were combined into a single physical abuse dimension. W. This scale was constructed to reflect 4 dimensions of empathy. These are perspective-taking (cognitive empathy), fantasy, empathic concern (emotional empathy), and personal distress. No items were rejected from either the perspective-taking or empathic concern subscales. One item was rejected from the fantasy subscale and 2 were rejected from personal distress on the basis of both internal and external consistency. The final perspective-taking, fantasy, and empathic concern subscales ran parallel to one another with respect to outside variables. As such, they were combined into a measure of global empathy. The_1nst_flgrld_59a1e. For this measure of just world beliefs, no items were rejected on the basis of either criterion of unidimensionality. 61 The_1,_2‘_and_g_5931g§. This locus of control measure was constructed so as to reflect the 3 dimensions of internality, powerful others, and chance. On the internality subscale, 3 items did not meet the criteria of unidimensionality. On the powerful others subscale, 1 item was rejected. No items were rejected from the chance subscale. The three subscales demonstrated parallelism to one another (after the I scale was reverse scored). As such, they were combined into a single external locus of control measure. Ihg_sg1f;ng:gggtign_lng§x_Lfinll. This questionnaire was assumed to reflect four dimensions. These are childhood self-derogation, childhood self-blame, current derogation of childhood self, and current blame of childhood self. For each of these four subscales, no items were rejected on the basis of either internal or external consistency. WW- (1) For each of the 8 scenarios, 4 items measured personal similarity to the particular child victim. Each of the 8 personal similarity subscales demonstrated structural unidimensionality. In addition, the 8 subscales were parallel to one another. As such, they were combined into a single personal similarity scale. 62 Table 4 Items from the Behavioral Child Blame Scale and their Factor Loadings onto the three Dimensions of Child Blame Child Blame Dimension * WW 6. 7. 10. 11. 12. 13. 14. 15. Jonathan was responsible for what happened to him because of what he did or said. Jonathan got what he deserved because of what he did or said. The events of that day were Jonathan's fault because of what he did or said. Jonathan is to blame for what happened to him because of what he did or said. Janice was responsible for what happened to her because of what she did or said. Janice got what she deserved because of what she did or said. The events of that day were Janice’s fault because of what she did or said. Janice is to blame for what happened to her because of what she did or said. Billy was responsible for what happened to him because of what he did or said. Billy got what he deserved because of what he did or said. The events of that day were Billy’s fault because of what he did or said. Billy is to blame for what happened to him because of what he did or said. Rachel was responsible for what happened to her because of what she did or said. Rachel got what she deserved because of what she did or said. The events of that day were Rachel's fault because of what she did or said. 1 2 3 .72 .58 .49 .70 .60 .59 .73 .62 .51 .74 .63 .54 .76 .59 .53 .75 .63 .67 .76 .63 .51 .80 .68 .58 .78 .62 .53 .75 .65 .69 .80 .68 .56 .80 .70 .60 .75 .60 .51 .75 .66 .65 .79 .67 .54 63 Table 4 (cont'd.) 16. Rachel is to blame for what happened to her because of what she did or said. 17. Patrick was responsible for what happened to him because of what he did or said. 18. Patrick got what he deserved because of what he did or said. 19. The events of that day were Patrick’s fault because of what he did or said. 20. Patrick is to blame for what happened to him because of what he did or said. 21. Phillip was responsible for what happened to him because of what he did or said. 22. Phillip got what he deserved because of what he did or said. 23. The events of that day were Phillip’s fault because of what he did or said. 24. Phillip is to blame for what happened to him because of what he did or said. 25. Jessica was responsible for what happened to her because of what she did or said. 26. Jessica got what she deserved because of what she did or said. 27. The events of that day were Jessica's fault because of what she did or said. 28. Jessica is to blame for what happened to her because of what she did or said. 29. Catherine was responsible for what happened to her because of what she did or said. 30. Catherine got what she deserved because of what she did or said. 31. The events of that day were Catherine's fault because of what she did or said. 32. Catherine is to blame for what happened to her because of what she did or said. .82 .70 .65 .69 .72 .65 .59 .63 .64 .65 .60 .60 .63 .69 .67 .72 .73 .71 .56 .55 .61 .61 .55 .57 .57 .59 .56 .56 .53 .59 .55 .58 .61 .64 .60 .51 .58 .50 .53 .47 .50 .45 .44 .47 .53 .40 .49 .44 .58 .47 .50 * 1 - behavioral child blame: 2 - characterological child blame: 3 - abuser justification 64 (2) For each of the 8 scenarios, 4 items measured behavioral child blame. Each of the 8 behavioral child blame subscales demonstrated both internal and external consistency. Moreover, the 8 subscales were parallel to one another. Consequently, they were combined into a global behavioral child blame scale. The factor loadings for all items of the behavioral child blame scale onto the behavioral child blame dimension ranged from .59 to .82. Table 4 demonstrates the factor loadings of each of the items from the behavioral child blame scale onto the 3 dimensions of child blame: behavioral, characterological, and abuser justification (all 95 < .001). (3) For each of the 8 scenarios, 4 items measured characterological child blame. Each of the 8 characterological child blame subscales demonstrated both internal and external consistency. Moreover, the 8 subscales were parallel to one another. Consequently, they were combined into a global characterological child blame scale. The factor loadings for all items of the characterological child blame scale onto the characterological child blame dimension ranged from .66 to .83. Table 5 demonstrates the factor loadings of each of the items from the characterological child blame scale onto the 3 dimensions of child blame (all ps < .001). 65 Table 5 Items from the Characterological Child Blame Scale and their Factor Loadings onto the three Dimensions of Child Blame Child Blame Dimension * 1 2 3 We 1. Jonathan was responsible for .63 .70 .45 what happened to him because of the kind of person he is. 2. Jonathan got what he deserved .61 .71 .50 because of the kind of person he is. 3. The events of that day were .59 .70 .43 Jonathan's fault because of the kind of person he is. 4. Jonathan is to blame for .61 .71 .47 what happened to him because of the kind of person he is. 5. Janice was responsible for .70 .80 .50 what happened to her because of the kind of person she is. 6. Janice got what she deserved .69 .79 .57 because of the kind of person she is. 7. The events of that day were .68 .80 .50 Janice's fault because of the kind of person she is. 8. Janice is to blame for .71 .82 .52 what happened to her because of the kind of person she is. 9. Billy was responsible for .71 .80 .53 what happened to him because of the kind of person he is. 10. Billy got what he deserved .68 .77 .59 because of the kind of person he is. 11. The events of that day were .67 .79 .50 Billy's fault because of the kind of person he is. 12. Billy is to blame for .70 .81 .56 what happened to him because of the kind of person he is. 13. Rachel was responsible for .70 .81 .53 what happened to her because of the kind of person she is. 14. Rachel got what she deserved .71 .79 .61 because of the kind of person she is. 15. The events of that day were .68 .80 .52 Rachel’s fault because of the kind of person she is. 66 Table 5 (cont'd.) 16. Rachel is to blame for what happened to her because of the kind of person she is. 17. Patrick was responsible for what happened to him because of the kind of person he is. 18. Patrick got what he deserved because of the kind of person he is. 19. The events of that day were Patrick's fault because of the kind of person he is. 20. Patrick is to blame for what happened to him because of the kind of person he is. 21. Phillip was responsible for what happened to him because of the kind of person he is. 22. Phillip got what he deserved because of the kind of person he is. 23. The events of that day were Phillip's fault because of the kind of person he is. 24. Phillip is to blame for what happened to him because of the kind of person he is. 25. Jessica was responsible for what happened to her because of the kind of person she is. 26. Jessica got what she deserved because of the kind of person she is. 27. The events of that day were Jessica’s fault because of the kind of person she is. 28. Jessica is to blame for what happened to her because of the kind of person she is. 29. Catherine was responsible for what happened to her because of the kind of person she is. 30. Catherine got what she deserved because of the kind of person she is. 31. The events of that day were Catherine's fault because of the kind of person she is. 32. Catherine is to blame for what happened to her because of the kind of person she is. .71 .60 .61 .58 .61 .57 .56 .55 .58 .58 .57 .56 .57 .64 .64 .63 .66 .83 .70 .70 .71 .73 .67 .66 .67 .69 .68 .68 .69 .68 .75 .72 .77 .55 .46 .49 .43 .48 .43 .45 .42 .43 .42 .46 .42 .42 .44 .51 .46 .49 * 1 - behavioral child blame; 2 = characterological child blame: 3 - abuser justification 67 (4) For each of the 8 scenarios, 4 items measured abuser justification. From each of the 8 abuser justification subscales, 1 item was rejected on the basis of both internal and external consistency. The rejected item read: "Most people would have responded as this parent did." Once this item was eliminated from each of the 8 subscales, all the final subscales demonstrated both internal and external consistency. In addition, since these 8 subscales were parallel to one another, they were combined into a global abuser justification scale. The factor loadings for all items of the abuser justification scale onto the abuser justification dimension ranged from .53 to .76. Table 6 demonstrates the factor loadings of each of the items from the abuser justification scale onto the 3 dimensions of child blame (all as < .001). 68 Table 6 Items from the Abuser Justification Scale and their Factor Loadings onto the three Dimensions of Child Blame Child Blame Dimension * MW 1. 4. S. 6. 7. 9. 10. 11. 12. 13. 14. 15. Sam Krazdin’s behavioral response toward Jonathan was justified. Sam Krazdin’s actions toward Jonathan were appropriate to what Jonathan did. Sam Krazdin really had no alternatives other than what he did. David Wood's behavioral response toward Janice was justified. David Wood’s actions toward Janice were appropriate to what Janice did. David Wood really had no alternatives other than what he did. Sandra Shale's behavioral response toward Billy was justified. Sandra Shale's actions toward Billy were appropriate to what Billy did. Sandra Shale really had no alternatives other than what she did. Judith Mills' behavioral response toward Rachel was justified. Judith Mills’ actions toward Rachel were appropriate to what Rachel did. Judith Mills really had no alternatives other than what she did. William Spence's behavioral response toward Patrick was justified. William Spence’s actions toward Patrick were appropriate to what Patrick did. William Spence really had no alternatives other than what he did. 1 2 3 .57 .48 .62 .55 .49 .61 .47 .45 .64 .61 .51 .74 .59 .51 .75 .46 .43 .72 .64 .54 .73 .63 .57 .74 .50 .49 .71 .61 .52 .73 .62 .57 .76 .49 .51 .70 .53 .43 .65 .52 .45 .67 .44 .40 .65 69 Table 6 (cont’d.) 16. Veronica Weiner's behavioral response .41 .32 .56 toward Phillip was justified. 17. Veronica Weiner's actions toward .39 .33 .56 Phillip were appropriate to what Phillip did. 18. Veronica Weiner really had no .33 .35 .54 alternatives other than what she did. 19. Harold Bateman's behavioral response .46 .37 .58 toward Jessica was justified. 20. Harold Bateman's actions toward .45 .39 .61 Jessica were appropriate to what Jessica did. 21. Harold Bateman really had no .34 .35 .53 alternatives other than what he did. 22. Helen Willis' behavioral response .54 .43 .68 toward Catherine was justified. 23. Helen Willis’ actions toward .53 .45 .67 Catherine were appropriate to what Catherine did. 24. Helen Willis really had no .37 .34 .57 alternatives other than what she did. * 1 - behavioral child blame: 2 = characterological child blame: 3 = abuser justification W. Hypothesis 5 asserted that behavioral and characterological components of victim blame, as well as abuser justification would all emerge as distinct factors. In order to test this hypothesis, a second-order confirmatory factor analysis was conducted. In this procedure, the 3 child blame scales were treated as 3 items. The results indicated that the scales met the second-order criteria of both internal consistency as well as parallelism. This suggests that the three components of child blame are specific factors of a higher order child blame factor. §§§g§§§>§834 888831836: 85:25 28: 8585s.... 7.2% 32% ”8.55822; 8: $2.: EagEsEsa €2.23 1 / X... \c\ e. s. s. 2.55 as. use: 25 2.. gas as; 70 Table 7 demonstrates the correlations of the three relevant scales with outside constructs. One variable in particular, personal similarity, was hypothesized to correlate with behavioral and characterological components of child blame in opposite directions. However, the results indicated that personal similarity correlated virtually identically with each of the 3 components of child blame. The three child blame scales loaded onto the global child blame factor as follows: .97, .88, and .79 for behavioral, characterological, and abuser justification respectively (see Figure 3). It is instructive to note that mean scores on behavioral child blame were significantly higher than mean scores on characterological child blame for subjects low on personal similarity (F (1, 458) = 403.33, p < .001) and for subjects high on personal similarity (F (1, 382) a 318.18, p < .001). Thus on average, subjects ascribed more behavioral than characterological blame. MW- 0n the basis of content, four subscales were derived from the ARVS. These were items that reflected the beliefs that rape victims were: (a) lying: (b) overly provocative: (c) sexually desiring rape; and (d) sexually experienced. (1) The factor loadings for all items of the lies scale onto the lies dimension ranged from .54 to .74: except for 1 item for which the factor loading was .24. Table 8 71 demonstrates the factor loadings of each of the items from the lies scale onto the 4 dimensions of rape victim blame: lies, provocativeness, sexual desire, and sexual experience (all us < .001). 72 Table 7 Correlations of the three Dimensions of Child Blame with Outside Constructs Child Blame Dimension * W 1 2 3 Subject Gender -.21 -.23 -.22 Subject Age -.07 -.05 -.03 Socio-economic Status -.01 .04 -.06 Physical Abuse (Father) .11 .06 .11 Physical Abuse (Mother) .18 .13 .22 Psychological Aggression (Father) .05 .04 .02 Psychological Aggression (Mother) .08 .04 .10 Neglect (Father) .06 .04 .07 Neglect (Mother) .11 .01 .13 Perspective Taking Empathy -.22 -.16 -.18 Fantasy Empathy -.23 -.21 .23 Empathic Concern -.23 -.22 -.29 Personal Distress Empathy .07 .08 .12 Just World -.02 .01 -.05 Internal Locus of Control -.14 -.09 -.14 Powerful Others Locus of Control .25 .21 .25 Chance Locus of Control .26 .23 .21 Personal Similarity .16 .17 .16 Childhood Self-Derogation .09 .05 .10 Childhood Self-Blame -.01 .02 -.05 Current Derogation of ' .13 .11 .18 Childhood Self Current Blame of .13 .13 .10 Childhood Self Rape Victim Blame .43 .44 .40 * 1 - behavioral child blame: 2 = characterological child blame; 3 s abuser justification 73 Table 8 Items from the Rape Victim Blame Scales and their Factor Loadings onto the four Dimensions of Rape Victim Blame Rape Victim Blame Dimension * 1 2 3 4 Items from the Lies Dimension of Rape Victim Blame 1. Women often claim rape to .66 .32 .44 .37 protect their reputations. 2. Even women who feel guilty about .24 .37 .12 .18 engaging in premarital sex are not likely to falsely claim rape (-). 3. Many women invent rape stories .74 .41 .44 .34 if they learn they are pregnant. 4. Many women claim rape if they have .70 .38 .41 .29 consented to sexual relations but have changed their minds afterward. 5. Accusations of rape by bar girls, .54 .41 .48 .33 dance hostesses and prostitutes should be viewed with suspicion. 6. Many women who report rape are .59 .34 .53 .41 lying because they are angry or want to get revenge on the accused. Items from the Pzgyggatiygngss Dimension of Rape Victim Blame 1. Women do not provoke rape by their .40 .64 .29 .24 appearance or behavior (~). 2. Men, not women, are responsible .31 .42 .23 .23 for rape (-). 3. WOmen who wear short skirts or .37 .62 .30 .20 tight blouses are not inviting rape (-). 74 Table 8 (cont'd.) Items from the ngnn1_n§§ize Dimension of Rape Victim Blame 1. It would do some women good .33 .21 .52 .56 to be raped. 2. Most women secretly desire .39 .30 .70 .52 to be raped. 3. Women put themselves in situations .53 .37 .57 .64 in which they are likely to be sexually assaulted because they have an unconscious wish to be raped. Items from the Sexual_£xpgzigngg Dimension of Rape Victim Blame 1. Women who have had prior sexual .28 .21 .44 .33 relationships should not complain about rape. 2. Sexually experienced women are not .36 .24 .62 .70 really damaged by rape. 3. In most cases when a woman was .29 .24 .58 .67 raped, she deserved it. * 1 - lies: 2 - provocativeness: 3 - sexual desire: 4 = sexual experience (2) The factor loadings for all items of the provocativeness scale onto the provocativeness dimension ranged from .42 to .64. Table 8 demonstrates the factor loadings of each of the items from the provocativeness scale onto the 4 dimensions of rape victim blame (all ye < .001). (3) The factor loadings for all items of the sexual desire scale onto the sexual desire dimension ranged from .52 to .70. Table 8 demonstrates the factor loadings of 75 each of the items from the sexual desire scale onto the 4 dimensions of rape victim blame (all ps < .001). (4) The factor loadings for all items of the sexual experience scale onto the sexual experience dimension ranged from .33 to .70. Table 8 demonstrates the factor loadings of each of the items from the sexual experience scale onto the 4 dimensions of rape victim blame (all us < .001). (5) A further test of the dimensionality of the ARVS was carried out. That is, in order to test the possibility that the four rape victim blame scales (lies, provocativeness, sexual desire, and sexual experience) are specific factors of a global rape victim blame dimension, a second-order confirmatory factor analysis was conducted. The 4 rape victim blame scales were treated as 4 items. The results indicated that the scales met the second-order criteria of both internal consistency as well as parallelism. Table 9 demonstrates the correlations of the four relevant scales with outside constructs. These results indicate a higher order rape victim blame factor. 76 Table 9 Correlations of the four Dimensions of Rape Victim Blame with Outside Constructs Rape Victim Blame Dimension * 1 2 3 4 We; Subject Gender -.24 -.27 -.33 -.24 Socio-economic Status .02 -.03 .02 .05 Physical Abuse (Father) .15 .09 .26 .13 Physical Abuse (Mother) .11 .06 .12 .12 Psychological Aggression .07 .00 .13 .05 (Father) Psychological Aggression .06 -.02 .09 .06 (Mother) Neglect (Father) .01 -.04 .13 .08 Neglect (Mother) .14 .07 .08 .09 Perspective Taking Empathy -.20 -.06 -.19 -.05 Fantasy Empathy -.20 -.26 -.13 -.13 Empathic Concern -.22 -.15 -.25 -.17 Personal Distress Empathy .13 .02 .16 .13 Internal Locus of Control -.24 -.15 -.20 -.14 Powerful Others Loc. of C. .25 .18 .23 .10 Chance Locus of Control .33 .11 .25 .10 Personal Similarity .10 .01 .26 .20 Childhood Self-Derogation .08 .02 .24 .19 Childhood Self-Blame .01 .02 -.03 -.04 Current Derogation of .14 .03 .26 .22 Childhood Self Current Blame of .06 .08 -.02 -.05 Childhood Self Child Blame .37 .44 .33 .33 * 1 - lies: 2 - provocativeness: 3 = sexual desire: 4 8 sexual experience 77 The four rape victim blame scales loaded onto the global rape victim blame factor as follows: .80, .60, .97, and .81 for lies, provocativeness, sexual desire, and sexual experience respectively (see Figure 3). The_CAsQ_and_fhs_ABXS_T9sstheri__A_Test_for_a_§lehal WWW. Hypothesis 6 asserted that blame directed toward rape victims and blame directed toward victims of child abuse would emerge as separate and distinguishable constructs. This hypothesis was tested as follows. The child blame factor and the rape victim blame factor were tested for parallelism to one another. Table 10 demonstrates the correlations of the two relevant factors with outside constructs. These results demonstrate parallelism between the two dimensions of victim blame indicating a still higher order victim blame factor. The correlation between the two dimensions of victim blame was .48. Thus, child blame and rape victim blame are estimated to load onto the global victim blame factor at 1,: .70 (see Figure 3). 78 Table 10 Correlations of the two Dimensions of Victim Blame with Outside Constructs Victim Blame Dimension * 1 2 We; Subject Gender -.25 -.34 Subject Age -.06 -.12 Socio-economic Status -.01 .02 Physical Abuse (Father) .11 .20 Physical Abuse (Mother) .20 .13 Psychological Aggression .05 .08 (Father) Psychological Aggression .09 .06 (Mother) Neglect (Father) .07 .06 Neglect (Mother) .09 .12 Perspective Taking Empathy -.21 -.16 Fantasy Empathy -.26 -.23 ' Empathic Concern -.28 -.25 Personal Distress Empathy .10 .14 Just WOrld -.02 -.13 Internal Locus of Control -.14 -.23 Powerful Others Loc. of C. .27 .24 Chance Locus of Control .27 .25 Personal Similarity .19 .18 Childhood Self-Derogation .09 .17 Childhood Self-Blame -.01 -.01 Current Derogation of .16 .21 Childhood Self Current Blame of .14 .02 Childhood Self * 1 - child blame: 2 = rape victim blame 79 82118111118135 Every relevant construct in the current research was estimated by means of multiple indicators (items). Consequently, for each of the respective variables, parallel-forms reliability coefficients could be estimated. The predominant method of estimation was the standardized Cronbach alpha coefficient (Cronbach, 1951). However, for all Guttman scales (i.e., the Conflict Tactics Scales), the odd-even split-half correlation (Spearman-Brown) was calculated. This latter procedure was necessary since Cronbach alpha coefficients tend to underestimate the reliability of Guttman scales. In addition to parallel-forms reliabilities, test- retest (Pearson correlation) coefficients were calculated for all scales administered twice. All reliability estimates are reported in Table 11. WW For each of the dimensions in the current study, mean scale scores were calculated. In addition, standard errors (of the means) and scale standard deviations were determined. These statistics are reported in Table 12. 80 Table 11 Parallel-Forms and Test-Retest Reliability Estimates P.F. T.R.* Measure (n - 866) (H - 126) Physical Abuse (Father) .76 Physical Abuse (Mother) .73 Total Physical Abuse .77 Psychological Aggression (Father) .82 Psychological Aggression (Mother) .83 Total Psychological Aggression .87 Neglect (Father) .53 Neglect (Mother) .48 Total Neglect .60 Perspective Taking (Cognitive Empathy) .79 Fantasy Empathy .78 Empathic Concern (Emotional Empathy) .78 Total Empathy .85 Just World .64 Internal Locus of Control .67 Powerful Others Locus of Control .75 Chance Locus of Control .70 Total External Locus of Control .83 Personal Similarity .98 .70 Childhood Self-Derogation .83 .58 Childhood Self-Blame .85 .61 Current Derogation of Childhood Self .87 .60 Current Blame of Childhood Self .93 .65 Behavioral Child Blame .97 .78 Characterological Child Blame .97 .78 Abuser Justification .95 .79 Total Child Blame .98 .82 Lies (Rape Victim Blame) .75 Provocativeness (Rape Victim Blame) .57 Sexual Desire (Rape Victim Blame) .62 Sexual Experience (Rape Victim Blame) .57 Total Rape Victim Blame .81 * Test-retest correlations available only for scales administered twice (i.e., scales from the CASQ and the SDI). N.B. For all reliability coefficients, as < .001. 81 Table 12 Scale Means, Standard Errors, and Standard Deviations Mean St. St. Error Dev. Subject Age 18.94 .05 1.50 Socio-economic Status 57.47 .66 19.24 Physical Abuse (Father) .19 .01 .17 Physical Abuse (Mother) .18 .01 .16 Total Physical Abuse .19 .01 .14 Psychological Aggression (Father) .54 .01 .31 Psychological Aggression (Mother) .47 .01 .31 Total Psychological Aggression .50 .01 .27 Neglect (Father) .21 .01 .26 Neglect (Mother) .19 .01 .24 Total Neglect .20 .01 .20 Total Empathy 2.70 .02 .52 Just World 4.22 .02 .60 Total External Locus of Control 2.53 .02 .61 Personal Similarity 2.21 .04 1.09 Childhood Self-Derogation 1.01 .03 .79 Childhood Self-Blame 1.96 .03 .98 Current Derogation of .88 .03 .85 Childhood Self Current Blame of 2.34 .04 1.15 Childhood Self Behavioral Child Blame 2.70 .04 1.11 Characterological Child Blame 2.17 .03 .95 Abuser Justification 1.70 .02 .72 Total Child Blame 2.21 .03 .87 Lies (Rape Victim-Blame) 1.30 .03 .74 Provocativeness (Rape 1.78 .03 1.00 Victim Blame) Sexual Desire (Rape .34 .02 .56 Victim Blame) Sexual Experience (Rape .17 .01 .41 Victim Blame) Total Rape Victim Blame .98 .02 .53 82 Study II: The Construct Dimensionality of Victim Blame Dissnssion The current study was oriented toward elucidating the nature of victim blame as a personality factor. Hypothesis 5 stated that behavioral, characterological, and abuser justification components of child blame would all emerge as distinct factors. The results did not support this proposition. Instead, the findings demonstrated that across a multitude of personality variables, behavioral child blame is parallel to characterological, and to abuser justification. This implies that all that differentiates the three scales is shared specific error (a kind of measurement error: Hunter 8 Gerbing, 1982). The three facets of child blame appear to be reflections of a single higher order child blame factor. Thus, there may be differences between individuals in the extent to which they blame abused children in general. However, speaking of a person's tendency toward behavioral versus characterological blame adds little additional information about that person. It should be noted that mean scores for behavioral child blame were higher than those for characterological. This finding suggests that as a group subjects ascribed more behavioral than characterological blame. However, group differences do not imply individual differences. These two kinds of blame do not act differently in distinguishing between persons. As an analogy, group temperatures measured 83 in Fahrenheit are higher than those measured in Celsius, even though differences between persons remain the same regardless of the measure used. At first glance, these results appear to contradict those of prior research. Janoff-Bulman (1979) distinguished between behavioral and characterological self-blame, and considered that distinction important. She concluded that characterological self-blame was typical of depressed persons, while behavioral self-blame was typical of rape victims. This one study has influenced the development of many research programs (e.g., Jensen 8 Gutek, 1982: Meyer 8 ”Taylor, 1986: Schoeneman, Hollis, Stevens, 8 Fischer, 1988: Thornton, 1984), as well as theoretical analyses (e.g., Shaver 8 Drown, 1986). However Janoff-Bulman's conclusions seem to have been inaccurate. Let us look at her depression data first. Janoff-Bulman found that depressed persons characterologically self-blamed more than nondepressed persons did. The significant F test she reported corresponds to a correlation coefficient of r - .19. She also found that depressed and nondepressed groups did not significantly differ on behavioral self-blame. The nonsignificant F test she reported here corresponds to a correlation coefficient of r - .14. A sampling error analysis of Janoff-Bulman's study indicates that any correlation below : - .17 would not have been "significant." 84 Thus, the difference in statistical significance between her two 2 values can be explained as a function of her sample size alone. Furthermore, is there really any meaningful difference (in terms of magnitude of effect) between a correlation of .19 and .14? Consequently, her conclusion that characterological, but not behavioral self-blame distinguishes depressed groups from nondepressed appears to be unwarranted. Next, Janoff-Bulman's (1979) rape data suggested that among rape victims who self-blamed, behavioral self-blame was more common than characterological self-blame. The major drawback to this finding is that she never determined whether there was a difference between behavioral and characterological blame for the depressed sample, nor for any other group. It may simply be the case that enyone who self-blames will be more likely to blame his/her own behaviors, rather than concluding that there is something wrong with his/her character. This would be consistent with the "blame avoidance motives" posited by defensive attribution theorists (e.g., Shaver, 1970). Thus, once again Janoff-Bulman's (1979) study does not allow for the conclusion that the behavioral/characterological distinction is one that is important in evaluating individual differences in personality. The findings of the current study seem to contradict the results of a second investigation as well. Thornton 85 (1984) found that in evaluating personally similar victims, behavioral blame was ascribed more than characterological blame while the reverse was true in evaluating personally dissimilar victims. At first glance, these results seem to demonstrate the importance of distinguishing between behavioral and characterological components of victim blame. However, the question Thornton should have asked was whether there were differences on each of the blame factors between enojeote who are personally similar versus personally dissimilar to the victim. In other words, in order to corroborate his model he should have found that as personal similarity increases, behavioral blame increases: and as personal similarity increases, characterological blame decreases. However, Thornton did not make these comparisons. In the current study it was demonstrated that the correlation between personal similarity and behavioral child blame was virtually identical to the correlation between personal similarity and characterological child blame. Once again, this finding is consistent with the position that behavioral and characterological components of victim blame do not distinguish subjects differently. Hyoozneeie_o posited that blame directed toward victims of child abuse and blame toward rape victims would also emerge as separate and distinguishable constructs. The results did not corroborate this assertion. It appears that child blame and rape blame are merely facets of a more 86 global victim blame personality factor. The current findings are not consistent with the suppositions made in prior research. Earlier investigators have assumed that victim blame in the case of rape is separate and distinct from victim blame in other areas. The research of Ward (1988), and Deitz et al. (1982, 1984) assumes that victim blame toward rape victims is a meaningful personality factor, and that such a construct as "rape empathy" exists. It seems more likely that these measures of rape victim blame are merely reflecting individual differences in global victim blame. It is possible that attribution of responsibility to victims may be a personality trait that differs in magnitude from person to person (of. Suedfeld, Hakstian, Rank, 8 Ballard, 1985). Recent arguments have been put forth (e.g., Funder, 1991) suggesting that global personality traits have far greater explanatory power than do narrow ones. Global traits relate specific behavioral observations to a complex and general pattern of behavior (Funder, 1991). Furthermore, general personality traits can be very stable across the lifespan (McCrae 8 Costa, 1984). It is important to note that the situations of child abuse and rape are both situations of interpersonal violence in which one individual is victimized by a powerful other. This limits the conclusions that may be drawn. Future research may better address the question of whether victim 87 blame is a global personality factor by utilizing very different victimization situations.‘ For example, it may be interesting to look at whether subjects who hold victims of rape responsible for the event, also hold poor persons responsible for their current socio-economic status. 88 Study III: The Path from Survivor of Abuse to Child Blame Results The confirmatory factor analyses yielded a factor by factor correlation matrix. All correlations were corrected for attenuation (of. Ghiselli, Campbell, 8 Zedeck, 1981). Table 13 presents the inter-factor correlation matrix which includes each of the relevant variables in this study. The correlations in this matrix were used to test hypotheses 7- 18. It should be noted that this matrix was generated separately for male and female subjects. The two resulting matrices were very similar to one another (to within sampling error) in terms of the magnitude of the correlations. It was concluded that gender did not act as a moderator variable for any of the comparisons presented below. 01‘ . ;— - ..:,’. .y; 5;, ,..;—. ..~ - .' .. . . : ...-. The results of the current investigation demonstrated a relationship between abuse as a child and child blame. The two variables correlated with one another at n - .18, o < .001. I§§§_91_H¥DQ§h§fi§S_Z:1& H!DQ§h§Si§.Z stated that adult survivors of physical abuse, in comparison to individuals having experienced less abuse, should demonstrate less cognitive empathy, and less emotional empathy. The confirmatory factor analyses 88A. .mo. v N .50.: v: m .3 :5. M H won .n... oo.» 3. no. 3. 3.: an. 3. a. 3.: 3.: on. 36:. I33» 2.3 33... .2 84 «a. an. 3.: 3. 3. an. 3.: 3.: 3. 363 63.0 «39—. .3 84 on. 3. no. 3.: 3.: I. no. 3.: :3 3283.8 3 3.3 2.380 .e co." «6.: 2.. «a. on. 2.: e1: «a. :3 36326 no 3332.3 2.08.36 .- ooé no. 3.: 3.: 3. ea. 3.: 363:3... 32.330 .6 co.“ an. an. 3.: 3.: on. 8339.3:uu3 36326 .e oo.« 3. 3.: 3.: 3. 33-2-3 3:63... .6. oo.« 5.: ani an. 35:60 no 303 3:83..- 33... .4 co; 3. 3.: 3388...... .n oo.« 3.: s53...» «36-. .« oo.u cease «cos-sac «each .u «H OH 0 e b o n v n a H 08.3.0: 58.2 8333.8 8883 E 88.. mH mHQme 89 indicated that cognitive and emotional empathy were part of a single dimension. As such, the two facets of empathy were combined. The correlation between total physical abuse and total empathy was r - -.13, p < .001. Hypothg§i§_§ asserted that individuals scoring lower on cognitive and emotional facets of empathy should score higher on child blame. The correlation between total empathy and total child blame was r a -.28, p < .001. Han§h§§i§_2 proposed that adult survivors of abuse, in comparison to individuals having experienced less abuse, should demonstrate lower levels of just world beliefs. The correlation between total physical abuse and just world beliefs was r = -.12, p < .001. flypgthgsis_lg stated that individuals indicating lower levels of just world beliefs, should score lower on child blame. The correlation between just world beliefs and total child blame was r - -.03, p > .05. The 95% confidence interval extended from -.10 to +.04. nggthgsis_11 asserted that adult survivors of abuse, in comparison to individuals having experienced less abuse, should demonstrate lower levels of internal locus of control, and higher levels of powerful others locus of control. The confirmatory factor analyses indicated that internal and powerful others locus of control scales were part of a single dimension. As such, the two facets were combined (after the internal scale was reverse scored) into 90 a single external locus of control dimension. The correlation between total physical abuse and total external locus of control was r - .21, p < .001. Hypothesis_lz proposed that individuals indicating lower levels of internal locus of control should score lower on behavioral child blame. nypgthgsis_11 stated that individuals demonstrating higher levels of powerful others locus of control should score lower on abuser justification and lower on behavioral child blame. As discussed above, the locus of control scales were combined into a total external locus of control dimension. In addition, the results from the confirmatory factor analyses led to the combining of the three facets of child blame, behavioral, characterological, and abuser justification into a total child blame factor. Thus, the correlation between total external locus of control and total child blame was r - .25, p < .001. Hypothg§1§_15 stated that adult survivors of abuse, to a greater extent than individuals having experienced less abuse, should see themselves as personally similar to abused children. The correlation between total physical abuse and personal similarity was r - .24, p < .001. H¥DQ§h£§1§_1§ asserted that individuals indicating higher perceptions of personal similarity to abused children should score higher on behavioral child blame, and lower on characterological child blame. The results indicated that 91 personal similarity correlated with behavioral, characterological, and abuser justification at rs - .16, .17, and .16 respectively (ps < .001). As discussed previously, all 3 components of child blame were demonstrated to be part of a single child blame dimension. The correlation between personal similarity and total child blame was z,- .17, p < .001. Hypg§h§§i§_1§ proposed that adult survivors of abuse, in comparison to individuals having experienced less abuse, should indicate having experienced during their childhoods, more self-derogation and more self-blame for parental punishment. The correlation between total physical abuse and childhood self-derogation was r - .36, p_< .001. The correlation between total physical abuse and childhood self- blame was r - -.07, p < .05. flypggng§i§_11 stated that individuals indicating more childhood experience of self-derogation and self-blame for parental punishment should demonstrate more current derogation of their childhood selves and more current self- blame for parental punishment. The correlation between childhood self-derogation and current derogation of childhood self was r - .74, n < .001. The correlation between childhood self-blame and current blame of childhood self was I - .48, p < .001. nypgtng§1§_1& asserted that individuals indicating more current derogation of their childhood selves and more 92 current self-blame for parental punishment should demonstrate more child blame. The correlation between current derogation of childhood self and total child blame was z,- .13, p < .001. The correlation between current blame of childhood self and child blame was r - .12, p < .001. W The results from the confirmatory factor analyses indicated that empathy, external locus of control, and child blame were each unidimensional constructs. These findings necessitated a change in the proposed path model. The model presented in Figure 1 was adjusted to the model presented in Figure 4, so as to reflect the unidimensional nature of the three variables discussed above. The model presented in Figure 4 was subjected to direct test using the technique of path analysis. The path coefficients were estimated using the traditional procedure of ordinary least squares. The program used was the routine PATH found in PACKAGE (Hunter & Cohen, 1969; Hunter, Gerbing, Cohen, & Nicol, 1980). This program generates path coefficients for recursive path models, when a correlation matrix of the relevant variables (factors) is inputed. All correlations should be corrected for attenuation. The correlation matrix presented in Table 13 was used as the input matrix for the path analyses. It is important 93 to note that when the linear effects due to socio-economic status were partialled out, the resulting matrix of partial correlations was virtually identical to the original matrix in Table 13. As such, SES appears to have had no systematic effect on these data. The results of the path analysis demonstrated that the model presented in Figure 4 was not consistent with the data. The Chi-square test for overall goodness of fit indicated a significant difference ( “K? (31) - 142.61, p < .05) between the model and the data. The correlations and the path coefficients for this path analysis are presented in Figure 5. In order to generate a model which more closely fit the data, the path model was modified somewhat. The changes in the model were guided by an attempt to reduce the size of the error terms (i.e., the differences between the actual and the reproduced correlation coefficients). The significant modifications were as follows. The self-blame variables (childhood and current) were removed; and just world was entered as an intervening variable between child physical abuse and external locus of control. Figure 6 represents the model that was generated. The results of the path analysis indicated that the Chi-square test for overall goodness of fit did not reveal a significant difference (‘x3 (14) - 19.13, p > .10) between 93A 33688348: . was: uBszBEBEagfia 8...ng $833 23...-.3 ‘Ill: 05...-..3 .55... 82.2.5 A S..&§..-..om =¢..au§..-..~m .55 AI-.. 322.5 \k\ ._.__.__.._____.H_.__.._. A A/ p A/ A/ , as... .82.... 2:... 322.2% .2329. ‘\ .228 .a was .25.... 28.. r .2: 52...... A If Huggflmfltmafiaommummag 5o. Hafifioams 5.53.. 33.1... 22.-.... null as... , 5...... ._ _- .555 7...... c8.......... A asses...” 5-..... .5 ..... 822.5 3.". T... :5 3...: .81.... .35 “she...” A .5... 2.... .22.... .c k «.83 3535 .3..&.E..-..~m .555 A 93B 2.5... 5%... 2.... ‘ .2.-. .u.-. a.-. .a.-. 93C ggsofisfimflogfiogfimmms "was... 5.3.3.2....8 3.8. .1 u . . .5..&.E..-..sm .555 ‘ .23... 5...... ._ _ EH. 3.5....” \\ .28.... A . W...“— 25... .56... + 2...... .5 .2. A/ > E. .n. .228 .c k / a... 5...... 2.5 as \ k 3...”.-. 2...— .n...e._ .2: a... ...... 94 the model and the data. This indicated a model that provided good fit for the data. It is instructive to note that both external locus of control and personal similarity predicted child blame. However, externality also predicted personal similarity. It was suggested that perhaps in addition to these linear effects, certain nonlinear effects may be present. That is, perhaps the correlation between external locus of control and child blame would be greater for subjects who score higher, compared to those who score lower, on personal similarity. The results indicated that such nonlinear interaction effects were not present. The correlations between external locus of control and child blame were rs = .26 and .22 (ps < .001) respectively, for subjects high and low on personal similarity. The difference between the two correlations is what would be expected from sampling error alone. 95 Study III: The Path from Survivor of Abuse to Child Blame Wu The correlational findings supported most of the original hypotheses of this investigation. The results from the path analyses indicated that the hypothesized path model was not perfectly consistent with the data. However, the path model that did provide good fit to the data, was very similar to the one hypothesized a priori. The following section will discuss the correlational and path analyses in terms of the original hypotheses. 0A; . ;; - '.,-. . .j- y“. ,._;; .; ., . . ._ . . ; ...-_ The findings indicated that individuals who reported having experienced more physical abuse as children, were more apt to blame child victims of abuse. Although no prior research has focused on this issue specifically, several earlier studies have been conducted on related issues. Libow and Doty (1979) demonstrated that more severely abused victims of rape tended to have the least punitive attitudes toward their assailants. In a somewhat related fashion to the current study, victims of rape tended to justify the abuser. In a study of child sexual abuse, Waterman and Foss-Goodman (1984) found that adult survivors of abuse were less likely to victim blame than controls. In contrast to the results of the current investigation, the results of the Waterman et al. study suggest that victims of (sexual) abuse are less prone to victim blame. However, in the Waterman et 96 al. investigation, it appears that subjects were asked if they were ever "molested" as a child. The use of such obvious language may have led subjects to respond in a "socially desirable" fashion. The current study, in part, was an investigation of the overall relationship between child physical abuse and child blame. However more important, this investigation was designed to determine the victim blame process. Recall that the path model that was proposed at the onset of this study suggested the possibility of antagonistic effects. Such effects were not found. Child physical abuse did not predict decreases in child blame through any of the intervening variables. In fact, through most of the intervening variables, child physical abuse predicted increases in child blame. In examining the relationship between child physical abuse and child blame, there are at least five important intervening variables. These variables will be discussed presently. These are: (a) empathy: (b) just world beliefs: (c) locus of control: (d) personal similarity: and (e) self- derogation. Empathy. The findings demonstrated that persons who have experienced higher levels of physical abuse, showed less disposition toward empathic response. This finding corroborated hypothesi§_z. Prior research has shown that abused children indicate lower levels of empathy than 97 nonabused children (Hain a George, 1985; Miller & Eisenberg, 1988: Straker & Jacobson, 1981). In addition, Koestner, Franz, and Weinberger (1990) found that individuals who had received more nurturant parental behaviors at a young age, demonstrated greater empathic skill later in life. The current study corroborated the results of Koestner et al. both in terms of direction and magnitude of effect sizes (to within sampling error). Straker and Jacobson suggested that an adult model of sharing, helping, and comforting is a major contributing factor in the development of empathy. A deprivation in such a model appears to lead to an impoverishment in empathic skill. Individuals demonstrating less empathy had a greater tendency to blame abused children. This finding corroborated hypothe§i§_§, The earlier studies in this area (Deitz, Blackwell, Daley, & Bentley, 1982; Deitz, Littman, & Bentley, 1984) suggested that individuals demonstrating low levels of rape empathy were more likely to blame victims of rape. A similar process appears to take place in the case of physical child abuse. The results from the path analysis suggest that individuals who experience physical abuse as a child become less empathic toward others. Furthermore, such decreases in empathic ability predict higher levels of child victim blame. 98 1nst_flnzld_ne11§fs. The attempt to view the world as a fair and just environment appeared to be an important factor in the current research. Hypothesi§_2 proposed that adult survivors of abuse should be less likely to view the world as a fair place. The results were consistent with this proposition. It appears that individuals who had experienced severely punitive childhoods have a greater tendency to assume that their actions will be met with unfair response. Earlier research in this field has demonstrated a variety of negative outcomes as a result of child physical abuse. These include deficits to cognitive ability (Vondra, Barnett, & Cicchetti, 1990), diminishment of self-concept and motivation (Baker 8 Baker, 1987), and increased aggression toward others (Briere a Runtz, 1990). However, prior research had not examined the effects of physical abuse on attributions of justice or fairness. The results of the current study imply that physical abuse may lead to increased suspiciousness of others' motives, and perhaps a sense that others have hurtful intentions. Lerner’s (1980) Just World hypothesis suggested that the assumption that world is fair, leads to the consequence of blaming victims for their misfortune (Rubin 8 Peplau, 1973: R. E. Smith, Heating, Hester, & Mitchell, 1976). In the current study, it was suggested that persons who believed the world was generally unfair would be less likely 99 to hold others responsible for their misfortune. Hypothesis 19 stated that individuals indicating lower levels of just world beliefs should score lower on child blame. The findings did not corroborate that assertion. The two variables did not significantly correlate with one another. It should be noted that the current study used a very large sample size. The confidence interval of the correlation between just world beliefs and child blame indicated that at most, the correlation between the 2 variables is very small in magnitude. These results are consistent with those of Kerr and Kurtz (1977) who found a nonsignificant near zero correlation between just world beliefs and victim blaming. In a similar vein, Ma and Smith (1985) discovered a near zero correlation between just world beliefs and authoritarianism. However, the current findings are not consistent with those of Rubin and Peplau (1973) as well as HacLean and Chown (1988) who found positive associations between believing in a just world and blaming victims. The current results may be better understood by making reference to another finding. Although just world beliefs did not correlate with child blame, they did predict locus of control. Individuals who endorsed beliefs in an unjust world, were more likely to have an external locus of control. This finding is consistent with several other studies in the area (Hotta 5 Tiegerman, 1979: Witt, 1988). 100 The current findings suggest that while beliefs in an unjust world may predict an external orientation, it may be this external locus of control orientation that is more important in predicting victim blame, than are the just world beliefs. The results of the path analysis were consistent with the position that individuals who experience physical abuse as a child are more likely to believe that the world is unjust. Persons who believe that they live in an unjust environment are more likely to assume that they do not control their own reinforcements. ngns_gf_ggn;rgl. Prior research in this area has suggested that abused children may differ from others in the belief in their ability to control their environment. Abused children tend to be more likely to attribute external control of reinforcement in their lives (Allen & Tarnowski, 1989: Barahal, Waterman, & Martin, 1981). Hypothesis_11 proposed that adult survivors of abuse, in comparison to individuals having experienced less abuse, should demonstrate lower levels of internal locus of control, and higher levels of powerful others locus of control. The findings provided corroboration for this assertion. It may be that the experience of physical abuse teaches children that they are not necessarily in control of physical consequences to their bodies. This belief in a lack of control appears to persist into adulthood. 101 Hypothe§i§_lz stated that individuals indicating lower levels of internal locus of control should score lower on behavioral child blame. Hypothesis_ll posited that individuals demonstrating higher levels of powerful others locus of control should score lower on abuser justification and lower on behavioral child blame. The findings indicated that in contrast to these predictions, total external locus of control predicted higher levels of child blame. The results were not consistent with the findings of several earlier studies (Paulsen, 1979; Phares & Lamiell, 1975; Phares & Wilson, 1972: Sosis, 1974; Thornton, Robbins, & Johnson, 1981: Zuckerman, Gerbasi, & Marion, 1977), which found lower levels of victim blame for externals. The apparent discrepancy between the current results and the prior literature may be explained as follows. First, many of the earlier studies that found internals to score higher on victim blame utilized single item measures of attribution of responsibility (e.g., Sosis, 1974; Zuckerman, Gerbasi, & Marion, 1977). In the current study, the reliability of the 96-item victim blame measure used was nearly perfect. Thus, measurement error was kept to a minimum. In addition, the large sample size meant that sampling error was also very low. Earlier studies did not go to such lengths to minimize measurement and sampling errors 0 102 Second, the current finding that externality of control is associated with victim blaming is consistent with a separate, but related literature in this area. Externality has been demonstrated to be positively associated with dogmatism (Russell & Jorgenson, 1978: Zuckerman, Gerbasi, 8 Marion), machiavellianism (Zuckerman & Gerbasi, 1977), and the tendency to use abusive conflict tactics with one's spouse (Rouse, 1984). All three of these variables should be associated with victim blame. In fact, Thornton, Ryckman, and Robbins (1982) found a strong positive relationship between dogmatism and blaming victims of sexual assault. Thus, there is some indirect support in the prior literature that external locus of control may predict higher levels of victim blame. The theoretical meaning behind the locus of control results may be best understood in light of the fact that the term locus of control really refers to locus of control of reinforcement. Phares and Lamiell (1975) stated that the concept deals with the extent to which individuals attribute the responsibility for the occurrence of reinforcement to themselves or to outside forces. As such, externals believe that they cannot make good things happen fgz_;hsnss1xss. Aronoff and Wilson (1985) viewed externality of control as an attribute of the "safety-oriented“ individual. Such persons are characterized by anxiety, insecurity, conformity, and dependence. Aronoff and Wilson argued that 103 in order to reduce the feelings of anxiety common in such persons, the motivation develops to structure experience in controllable and predictable ways. Consequently, such persons develop dogmatic or authoritarian belief systems. Blaming victims may provide a sense of personal control or predictability of outcome that such persons seek. The theoretical conceptualization regarding safety orientation developed by Aronoff and Wilson (1985) appears to be consistent with the current findings. The results of the path analysis suggest that individuals who have experienced physical abuse as a child (people who would be expected to have greater safety needs), go on to believe that they are not capable of bringing about their own reinforcements. Such persons are more likely to blame victims of physical child abuse. It may be that attributing responsibility to victims brings about greater perceived predictability of outcome and a sense of personal control. W. W stated that adult survivors of abuse should be more likely to view themselves as personally similar to abused children. The findings corroborated that position. Persons who had experienced physically punitive parenting were more likely to endorse items indicating greater identification or "oneness" with hypothetical abused children. This finding is particularly interesting in light of the observation that adult survivors of abuse demonstrated less overall empathic tendency. 104 Survivors of abuse may have the capacity to recognize that they have undergone similar life experiences that certain others have as well. However, this does not mean that they become more empathic persons. It appears that early deprivation of a nurturant model (Hoestner, Franz, and Weinberger, 1990: Straker & Jacobson, 1981) leads to subsequent deficits in empathic skill. It was proposed by Shaver (1970) that attributions of responsibility to victims will be influenced by two motives. Individuals have a need to defend against the possibility that random misfortune may happen to themselves (harm- avoidance). In addition, persons are motivated to defend against the possibility that they will be held responsible if they were to end in a similar fate (blame-avoidance). Thornton (1984) found that in evaluating personally similar victims, behavioral blame was ascribed more than characterological blame while the reverse was true in evaluating personally dissimilar victims. He explained this in terms of harm-avoidance and blame-avoidance motives. As discussed previously, Thornton should have looked at whether there were differences hstussn_snhjsshs personally similar versus personally dissimilar to victims on each of the behavioral and characterological components of victim blame. In the current investigation it was suggested (hypgghssis_1§) that individuals indicating higher perceptions of personal similarity to abused children should 105 score higher on behavioral child blame, and lower on characterological child blame. The findings did not corroborate that expectation. Personal similarity correlated positively with each of the 3 components of child blame. Furthermore, the results from the confirmatory factor analyses revealed that the 3 kinds of child blame were actually part of a single child blame dimension. When the relationship between personal similarity and global child blame was assessed, the findings indicated that persons high on personal similarity were more likely to blame abused children for the event. It appears that persons identifying with abused children respond by viewing such children as responsible. It may be that persons who consider themselves to be similar to abused children are individuals who have incorporated into their own identities the concept of victim. Such persons would have much to feel unsafe about. Harm- avoidance motives would be particularly salient in their lives. Thus, blaming child victims would be a natural response. However, the motivation to view such children as responsible may be so great that such children are blamed in a variety of ways, not just for their behaviors. That is, they are blamed for nhs_;hs¥_szs in addition to what they did. Furthermore, the abusive response of the parent is viewed as justified. 106 The results of the path analyses were consistent with the position that adult survivors of abuse view themselves as similar to children who have endured such physical punishment. Persons identifying with abused children have a greater tendency to attribute to them responsibility for their misfortune. aslf;nszsgstigh. Earlier research in this area suggested that abused children experience considerable self- derogation as a consequence of physical abuse. Theory and research has focused on the effects of physical abuse on self-esteem or self-image (Allen & Tarnowski, 1989; Green, 1982: Hjorth & Ostrov, 1982; Steele, 1986). Cohen (1984) suggested that the abused child self-derogates in order to justify the parent's behaviors, an act of family "loyalty." It has been suggested that in addition to derogating themselves, abused children actually blame themselves for the punishment they have endured (Green, 1982: Amsterdam, Brill, Bell, and Edwards, 1979). However, research has been somewhat inconsistent in this regard. Ney, Moore, MoPhee, & Trought (1986) found abused children to blame themselves in all but the most severe cases; but Herzberger, Potts, and Dillon (1981) did not find differences between abused and nonabused children in terms of self-blame. Hypgthssis_1§ stated that adult survivors of abuse, in comparison to individuals having experienced less abuse, should be more likely to indicate having experienced during 107 their childhoods, more self-derogation and more self-blame for parental punishment. The results only partially corroborated this hypothesis. Individuals who had experienced higher levels of physical abuse demonstrated higher levels of self-derogation, but (marginally) lower levels of self-blame. One interpretation for these results may be that although adult survivors of abuse believed they were naughty children, they also believed that they did not deserve to receive as harsh punishment as they received from their parents. This finding seems to run contrary to Cohen’s (1984) assertion that the abused child blames him/herself in order to justify the parent's behaviors. In this study, survivors of abuse did not justify their parents’ behaviors. That is, they were less inclined to believe that they deserved to receive the punishment their parents chose to give than were subjects who were not abused. If survivors of abuse do not blame themselves to justify their parents' actions, then the question remains as to why abused persons self-derogate (think of themselves as "bad"). It may be that through a social modelling process, abused children learn that they are naughty children. Physical beatings associated with vituperative statements from the parents convey the message that the child is a bad person. Repeated exposure to this notion may eventually become incorporated into the child's own self-image. 108 Hypgthssis_lz suggested that individuals indicating more childhood experience of self-derogation and self-blame for parental punishment should demonstrate more current derogation of their childhood selves and more current self- blame for parental punishment. The findings corroborated this assertion. One interpretation for these findings is that the way in which subjects viewed themselves in childhood persisted into adulthood. However, another equally plausible interpretation may be that subjects had difficulty differentiating between the two general time frames involved. That is, it may have been difficult for subjects to recall their childhood self-concepts as distinct from their current judgment of what they were like as children. Thus, the high correlations may be an artifact of the failure to discriminate between the two time frames. flypghhssis_ls stated that persons who indicated more current derogation of their childhood selves and more current self-blame for parental punishment should demonstrate more child blame. The findings were consistent with that proposition. It appears that subjects who view themselves negatively, and who consider themselves as having been deserving of punishment, believe that others too, are blameworthy. These persons seem to view others as negatively as they view themselves. The results from the path analyses are consistent with the position that individuals who have undergone physically 109 abusive parenting think that when they were children, they_ viewed themselves as bad or naughty. Such persons believe that, in fact, they were naughty children. Individuals viewing themselves as having been "bad children" view others in a similar fashion. Hence they are more likely to child blame. W The results of this study imply several possible intervening processes in the relationship between physical child abuse and child victim blame. These processes are presented as follows. (1) The experience of physical abuse leads one to take on a particular attitude toward him/herself. It appears that harsh physical beatings communicate a powerful message to an individual regarding the kind of person s/he is. That is, that the child is naughty, disobedient, and in general a bad person. Viewing others as we view ourselves, the potential consequence is that other child victims also are seen as responsible for the suffering they endure. The experience of child physical abuse may also lead persons to incorporate into their identities the belief that they are “victims." When observing other similar victims, such persons may feel considerable threat that a similar fate may befall them too. In the interest of avoiding perceived harm, child victims are viewed as responsible for their situation. 110 (2) The experience of abuse affects the way one learns to relate to others. In addition to influencing the particular view one takes of the self, harsh physical abuse also teaches a style of interpersonal contact. Repeated physical punishment that may or may not be necessary, implicitly conveys the notion that understanding others' feelings is unnecessary. Furthermore, the abused child's deficits in empathic skill may also be a function of what s/he is not exposed to; namely, a model of warm caring relations. Consequently, persons who fail to develop empathy, may not have the ability to comprehend the suffering of others. They may be more apt to view other child victims in a negative light. (3) The experience of physical abuse leads to a particular set of assumptions regarding the nature of one's environment. Individuals who have endured harsh physical punishment learn that their world is quite hostile. Events are construed as unfair. Furthermore, they learn that they are not in control of reinforcement in their lives. Such a view may create a certain element of uncertainty of outcome. In the interest of avoiding the threatening possibility that unpredictable and potentially harmful outcome may befall them, it is comforting to view those who suffer child maltreatment as having brought about their own painful destinies. 111 MW One limitation of the current study is that subjects were divided into childhood experience of abuse based on retrospective self-report. It is possible that some subjects who were physically abused as children, may have consciously or unconsciously denied their punitive experience in the current investigation. However, three reasons may be given as to why the self-report classification may still have been accurate. First, the percentage of subjects in this study who did endorse levels parental punitiveness that may be considered abusive, was comparable to the percentage of persons in the population estimated to have been abused (Azar & Wolfe, 1989: Helfer, 1987). Second, subjects in this study were never asked whether they were "abused." A study by Rausch and Knutson (1991) suggested that only 26.6% of subjects classified as having been abused, considered themselves to have been abused. As such, subjects in the current research were asked to endorse specific parental behaviors, separately for each parent. Earlier studies have demonstrated that using the technique of specifying the punitive parental behavior can yield reliable and valid classifications of abuse. For example, the Assessing Environments Questionnaire (AEIII: Berger, Knutson, Mehm, & Perkins, 1988) has demonstrated good reliability and validity for detecting childhood experience of abuse. Third, many steps were taken to insure 112 subject anonymity of response. This anonymity was communicated clearly to the participants. A second limitation to this study relates to the question of causal inference. The current study was cross- sectional in design. Subjects were asked to report their childhood experiences of abusive parenting. They were also asked to indicate their current victim blaming attitudes toward abused children. The direction of causation follows a particular logical time sequence. Nevertheless, there is the possibility that with such a cross sectional design, assumptions made regarding the direction of causality, may be erroneous. Longitudinal research of individual difference factors appears to be the natural solution to the question of causal inference. One direction for future research may be to investigate the process of victim blame over time, in groups of young children who have or have not been physically abused recently. 113 Additional Issues Explored The Effects of Gender Concordance on Personal Similarity Results flynghhssis_12 asserted that individuals judging victims of their own gender (gender concordant), in comparison to those judging victims of the opposite gender (gender discordant) should indicate higher perceptions of personal similarity to abused children. This hypothesis was tested by means of the Analysis of Variance technique. It was expected that male subjects would rate male children higher than female children on personal similarity: while female subjects would rate female children higher than male children on personal similarity. The results demonstrated a significant (F (1, 819) = 8.58, p,< .003), but very small (eta - .02) subject gender by child gender interaction on ratings of personal similarity. Planned comparisons revealed that among male subjects, there was no significant difference in personal similarity ratings of male and female children. Female subjects rated female children significantly higher (F (1, 578) a 18.35, p < .001) than male children on personal similarity. However the effect size for this difference was small (eta = .06). 114 Additional Issues Explored The Effects of Gender Concordance on Personal Similarity 121mm Hxngzhssis_12 stated that individuals judging victims of their own gender (gender concordant), should be more likely to indicate perceptions of personal similarity to abused children. The results provided only partial corroboration for that assertion. To a small extent, females identified more with female victims (consistent with Skrypnek, 1980); while males did not identify more with male victims. These findings may be explained as follows. It is possible that there is a gender difference in the process of victim identification. The gender of the victim may be more important to females in determining feelings toward the victim. While it is possible that such a gender difference exists, it may be small in terms of magnitude of effect. 115 Post-Hoe Findings Results As demonstrated above, a significant correlation was calculated between total physical abuse and childhood self- derogation. In addition, a significant correlation (r - .15, p < .001) was found between total physical abuse and xsriabilitx in childhood self-derogation. The latter variable was calculated as: (x - 56)1 where X = childhood self-derogation scores; ‘3 = mean childhood self-derogation scores One interpretation of this finding is that a specific moderator variable is operating in the relationship between total physical abuse and childhood self-derogation. It was proposed that the social support gained by having had a sibling would operate as such a moderator. Figure 7 demonstrates a plot of mean childhood self-derogation scores for physical abuse by siblings. For the physical abuse dimension, subjects were divided into those who have or have not suffered at least one form of severe parental violence. For the siblings dimension, subjects were divided into those with or without at least one sibling. To test the possibility that physical abuse interacts with the absence of siblings (no siblings) in the elicitation of childhood self-derogation, a regression analysis was conducted. Childhood self-derogation was 115A soapmmoumouuamm noosoasno :6 mmcaansm one manna Hmosmsga to powwow use up magmas swag. mmsg< .32.... .25.... oz . _ azmnc. om. ..mn:. cm. .9. as. a - 8.. 8.... 0...... s.o .o -c z . .zm . on. .m:.. as... “x 23...... 9.02 . 8.”... on... o.. 115B cogs—mommadamm 805035 so coBomumbcH owes“. use Macadam 0: mayo: 6952 Hogans mo woodman one no ousmE Eu...- ....x. .. A . 2.25.... 5...... -_ _ .2 .5 . ....=..u.....= - .3. E: “was.“ .x...___.__._. .8. .3. c2 .8. E . A: 2.2 116 regressed onto abuse (A), no siblings (NS), and the A by NS product. The interaction effect corresponded to the beta weight of the product variable with childhood self- derogation (cf. Cohen, 1978: Evans, 1991). The results of this analysis are displayed in Figure 8. It is instructive to note that the beta weight between the product variable and childhood self-derogation was positive, although it was small (h - .06, p < .08). 117 Post-Hoc Findings Discussion A number of theorists have espoused the so-called ”buffering hypothesis” of social support. That is, if social support is low, stress is more likely to lead to disorder than if social support is high (Fleming & Baum, 1986). Social support is said to protect the individual from the illnesses that arise from environmental stressors. The hypothesis suggests an interaction between social support and stress on negative consequences. This paradigm has been corroborated in studies using such samples as large community groups (Wilcox, 1981), persons living near nuclear power stations (Fleming, Baum, Gisriel, & Gatchel, 1982), and individuals in the navy (Sarason, Sarason, Potter, & Antoni, 1985). Cohen, McGowan, Fooskas, and Rose (1984) found a stress-buffering effect of perceived, but not received social support. The current investigation revealed an interaction between not having grown up with siblings, and having undergone child abuse, on the extent of childhood self- derogation. The experience of abuse may be seen as a life stressor. Not having siblings in one's environment during the stressful period, may be conceptualized as an absence of social support. Increased levels of childhood self- derogation may be seen as negative psychological consequence. The current findings appear to be consistent 118 with the position that the effects of child abuse are particularly detrimental to persons who have not grown up with siblings. Having at least one sibling may function as a supportive stress ”buffer," another individual with whom the abused child may commiserate. In addition, having a sibling in such an environment may provide a means of social comparison. The abused child, perhaps exposed to the suffering of a sibling, may be less likely to view himself/herself as unusual or I'weird." It should be stressed that these results were not hypothesized s priszi. As such, there is greater likelihood that random factors can account for the findings. These results should be viewed as exploratory and nondefinitive, rather than confirmatory. GENERAL DISCUSSION The current investigation was a study of victim blame in the case of child physical abuse. Several theorists (e.g., Endler, 1983) have suggested that in order to understand a particular psychological process, both situational as well as "person" (individual difference) factors must be analyzed. Some of the situational determinants of victim blame in child abuse were investigated in the current study. The results suggested that situational factors do appear to influence the extent of victim blame. For example, the genders of various participants in child abuse situations influence the attitudes that observers, in general, have toward such individuals. Future research may be directed toward understanding the role of a variety of other possible situational determinants of victim blame in child abuse. These include victim attractiveness, participant socio- economic status, and severity of abusive behavior. Prior research into victim blame processes have generally focused only on the effect of situational determinants in influencing attributions of responsibility. The current study suggests that investigating individual differences in such attributions may also be a fruitful course of inquiry. In addition to investigating situational 119 120 determinants of victim blame, the current study investigated "person" factors in the manifestation of victim blame. The results of this study were consistent with the position that victim blame is a global personality factor. Persons who blame victims in one domain of conflict do so in others as well. It appears that individuals differ in the extent to which they attribute responsibility to victims. The current study demonstrated several ways in which individual difference factors are associated with one another in the child blame process. In particular this investigation suggested that there may be three routes from having experienced physical abuse to having harsher attitudes toward abused children. The experience of abuse predicts changes: (a) in the individual's view of the self: (b) in the way the person relates to others: and (c) in assumptions made about the nature of the person's environment. These appear to converge in leading to greater blame of abused children. The finding that there are differences in victim blame attitudes among people, depending on whether or not they have themselves been abused, leads to an interesting theoretical question that only future research can answer. That is, are there individual differences in victim blame among persons who have abused their own children? If certain child abusers do view the child as generally responsible for the abuse, such individuals may experience 121 their own behaviors as ego-syntonic. On the other hand, if certain abusers are apt to view the child as blameless, the implication is that they experience their own abusive behaviors as ego-dystonic. It is possible that the factors influencing victim blame among child abusers are the same variables that influence victim blame in the general population. That is, for current child abusers, the extent to which they had experienced abuse as a child may influence the extent of current victim blame, leading to differences in the way they now experience their own behaviors. The area of child abuse has focused much research effort on the way in which child abusers differ from nonabusive, effective parents. A host of studies have demonstrated that child abusers, as a group, have lower self-esteem (Anderson & Lauderdale, 1982), experience considerable life stress (S. L. Smith, 1984: Wolfe, 1985), and have fewer social support networks (Gaudin E Pollane, 1983). However, research in this area has failed to investigate individual differences among abusers. These include both differences in the way they attribute responsibility to their children, as well as differences among those abusers who themselves were or were not abused as children. 122 WW5 Several theorists (e.g., Christensen, 1985: Kazdin, 1980) have discussed extensively the necessity for external validation of research findings. It has been suggested that the results of psychological research may not always be generalizable across persons, settings, or times. (1) The current research may have been somewhat limited in terms of its generalizability across persons. The sample size, although large, consisted only of college undergraduates. It may be that this group exhibited somewhat unique characteristics such as higher levels of intelligence, more education, and greater need for achievement. However, it is instructive to note that the socio-economic background of the average participant in the current study was only slightly higher than that of the average American, as measured on the Duncan SEI. One direction for future research may be to extend the current findings to samples other than college students. These samples may be parents in the community, children, and so on. It may be the case the current study made use of a groups of subjects that are somewhat homogeneous in terms of intelligence. Furthermore, as a group, they may have been brighter than the average American. However, it should be recalled that one of the outcomes of homogeneity of sample is to decrease, not increase the size of correlation 123 coefficients. When the standard deviations of the sample are restricted, so too are the magnitude of correlations. In addition, mean scores are of no consequence when calculating correlations. Thus, if the mean intelligence level of a group of subjects is higher than average, that will not affect the correlation coefficients, so long as the standard deviation is unaffected. In the context of the current study, these statistical issues suggest that if an attempt were made to replicate these findings on a community sample, the size of the correlations should turn out larger, not smaller. (2) The results also may be limited in terms of their generalizability across settings. The data were gathered by means of questionnaires that were filled out in college classrooms. Subjects were not observed in their natural environments. Researchers such as Tunnell (1977) have advocated the use of field settings in psychological research (as opposed to laboratory environments) due to their greater naturalness, and hence their ability to provide superior ecological validity. Another direction for future research may be to measure the relevant variables in the current study by means other than questionnaires, such as projective tests. Alternatively, subjects may be observed in natural environments. For example, the ways in which parents and children relate in real or "role-played" situations may be used to code scores on child blame. APPENDICES APPENDIX A CONFLICT TACTICS SCALES APPENDIX A CONFLICT TACTICS SCALES Think back to NW- Here is a list of things that your fathsr might have done when you had a conflict with him. Now taking into account all disagreements (not just the most serious one), we would like you to say how often he did the things listed. Answer by circling the correct response: A - Never D a 3-5 times G - 21-40 times B - Once E a 6-10 times H - 41-80 times C - Twice F 8 11-20 times I - > 80 times 1. Discussed an issue calmly..............A B C D E 2. Got information to back up his side of things.....................A 3. Brought in, or tried to bring in, someone to help settle things..........A 4. Ignored you for a day or two...........A 5. Ignored you for several weeks at a time..............................A 6. Would not make food or shelter or clothing available to you...........A 7. Insulted or swore at you...............A 8. Sulked or refused to talk about an issue.........................A 9. Stamped out of the room or house or yard.......................A 10. Cried..................................A 11. Did or said something to spite you.....A 12. Did or said something to hurt your feelings.....................A 13. Did or said something to embarrass or humiliate you.............A 14. Did or said something psychologically cruel to you...........A 15. Threatened to hit or throw something at you.................A 16. Threw or smashed or hit or kicked something................A 17. Threw something a; you.................A 18. Pushed, grabbed, or shoved you.........A l9. Slapped you or spanked you.............A 20. Kicked, bit, or hit you with a fist....A 21. Hit or tried to hit you with something.....................A 22. Beat you up............................A 23. Burned or scalded you..................A 24. Thzsshsnsd you with a knife or gun.....A 25. Used a knife or fired a gun............A U C D ”1"1’11'11'11 'li’ii'fl'd’fl ’1 '11 ’11 '11 "1'11"! "I "1"] '1! '11"! '11 '11 00000 00000 0 0 O 0 000 0 00 0 00 0 0 ”53:33:33 23:33:33: 3: :13 :3 a: 33:13:13 :3 33:13 33 5:11: 33 fl: HHHHH HHHHH H H H H HHH H HH H HH H H UWNUU woman: a: In (I! U GU03 U mu: G we: 00000 00000 0 0 0 0 000 0 00 0 00 UUUUU 00050 O U U 0 COO U DO U DO NMMMFJ PIPING!!! M M M M MM!!! M MEI M MD! I"! 126 APPENDIX B SELF-DEROGATION INDEX APPENDIX B SELF-DEROGATION INDEX Please answer these questions using the following scale: 0 - not at all 3 - fairly l - mildly 4 8 considerably 2 - moderately Think book to your childhood Wei- 1. 2. 8. 9. 10. 11. 12. When you were a child, to what extent did you see yourself as a bad child?...........0 When you were a child, to what extent did you see yourself as a naughty child?.......0 When you were a child, to what extent did you see yourself as a disobedient child?.......0 When your parents punished you as a child, to what extent did you blame yourself for what you got?...............0 When your parents punished you as a child, to what extent did you feel you got what you deserved?................O When your parents punished you as a child, to what extent did you feel responsible for what you got?.............0 To what extent do you now see yourself as having been a bad child?...........0 To what extent do you now see yourself as having been a naughty child?.......0 To what extent do you now see yourself as having been a disobedient child?............0 Thinking back to when your parents punished you, to what extent do you now blame yourself for what you got?...........o Thinking back to when your parents punished you, to what extent do you now feel you got what you deserved?... ...... ...0 Thinking back to when your parents punished you, to what extent do you now feel responsible for what you got?.........0 129 H F‘ to NNNN APPENDIX C CHILD ABUSE SCENARIOS QUESTIONNAIRE APPENDIX C CHILD ABUSE SCENARIOS QUESTIONNAIRE The following conflict situations are shortened versions of ag;na1_pnb11§hgd case studies (the names have been changed). Please read each of the situations carefully. Following the situations are a series of statements. You will probably agree with some items and disagree with others. We are interested in the extent to which you agree or disagree with these statements. Read each statement carefully. Then indicate the extent to which you agree or disagree by circling the number following each statement. There are no right or wrong answers. If you find that the numbers to be used in answering do not adequately reflect your own opinion, circle the one that is closest to the way you feel. Try to be as honest as possible. Thank you. Please write down today's date here: Your ID number is: 131 132 APPENDIX C (cont'd.) Jonathan Krazdin, an 8 year old boy was in his room loudly hitting his dog with a plastic baseball bat. The dog was in pain. Sam Krazdin, Jonathan’s father called several times for Jonathan to stop. Jonathan would not stop. He threw a toy at his father and yelled: "Make me!" Sam Krazdin pulled the plastic bat out of Jonathan's hands, and hit Jonathan with it several times. Jonathan later developed bruises. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 a strongly disagree 4 = neutral 2 = disagree 5 = agree somewhat 3 = disagree somewhat 6 = agree 7 = strongly agree 1. Jonathan's behavior was provocative (e.g., mischievous, insolent, or aggressive)......l 2 3 4 5 6 7 2. This event was severe and traumatic for Jonathan....... ...... ........1 2 3 4 5 6 7 3. Jonathan was responsible for what happened to him because of what he did or said.......... ..... ......l 2 3 4 5 6 7 4. Jonathan got what he deserved because of what he did or said.............1 2 3 4 5 6 7 5. The events of that day were Jonathan's fault because of what he did or said.....................1 2 3 4 5 6 7 6. Jonathan is to blame for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 7. Jonathan was responsible for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 8. Jonathan got what he deserved because of the kind of person he is........1 2 3 4 5 6 7 9. The events of that day were Jonathan’s fault because of the kind of person he is................1 2 3 4 5 6 7 10. Jonathan is to blame for what happened to him because of the kind of person he is................l 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 133 APPENDIX C (cont'd.) Sam Krazdin's behavioral response toward Jonathan was justified..............1 Sam Krazdin’s actions toward Jonathan were appropriate to what Jonathan did..........................1 Sam Krazdin really had no alternatives other than what he did........l Most people would have responded as Sam Krazdin did.........................1 I feel similar to Jonathan.................1 I identify with Jonathan...................1 Jonathan reminds me of myself..............1 I feel a sense of sameness with Jonathan..............................1 If the average parent had a child who did what Jonathan did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry 3::2::..... .ngry 3...:Ienraged NNNN N N N U h UUUU u huh-bub h h U" UIU'IUIU 0‘ 0| 0 0000 0 0 \lflfl‘l \l \J \l 134 APPENDIX C (cont'd.) David Wood came home from work. As he entered the house he noticed his income tax forms, which he had left in his desk, were sprawled throughout the living room. He called his 8 year old daughter, Janice, and firmly told her to clean up the mess she had made. She replied "I don't wanna. You do it." This went on for a few minutes until David removed a shoe and repeatedly hit Janice with it. Her left arm became bruised. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 = strongly disagree 4 = neutral 2 - disagree 5 = agree somewhat 3 8 disagree somewhat 6 = agree 7 - strongly agree 1. Janice's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Janice.......................1 2 3 4 5 6 7 3. Janice was responsible for what happened to her because of what she did or said....................1 2 3 4 5 6 7 4. Janice got what she deserved because of what she did or said............1 2 3 4 5 6 7 5. The events of that day were Janice's fault because of what she did or said....................1 2 3 4 5 6 7 6. Janice is to blame for what happened to her because of what she did or said....................1 2 3 4 5 6 7 7. Janice was responsible for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 8. Janice got what she deserved because of the kind of person she is.......1 2 3 4 S 6 7 9. The events of that day were Janice’s fault because of the kind of person she is...............1 2 3 4 5 6 7 10. Janice is to blame for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 135 APPENDIX C (cont’d.) David Wood's behavioral response toward Janice was justified................1 David Wood's actions toward Janice were appropriate to what Janice did............................1 David Wood really had no alternatives other than what he did........1 Most people would have responded as David Wood did..........................1 I feel similar to Janice...................1 I identify with Janice.....................1 Janice reminds me of myself................1 I feel a sense of sameness with Janice................................1 If the average parent had a child who did what Janice did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry g:::::quite angry 3...:Ienraged N NNNN N N UUUU U U uh ##bb U b 0 0000 0' U! 0 0000 0 0 Q QQQQ Q Q 136 APPENDIX C (cont'd.) Billy Shale, an 8 year old boy was playing in his room while his parents were sleeping. He pulled out his soccer ball and started kicking it loudly around the room, banging it off the walls. His mother, Sandra Shale, called for him to stop because she was trying to sleep. Billy ignored her and kept playing. She entered his room and said: "What did I tell you?" Billy kicked the ball at his mother and shouted: ”Leave me alone!" Sandra took a book off Billy's bookshelf and hit him hard several times. His back later became bruised. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 = strongly disagree 4 a neutral 2 - disagree 5 8 agree somewhat 3 8 disagree somewhat 6 2 agree 7 a strongly agree 1. Billy's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Billy.............. ..... .....1 2 3 4 5 6 7 3. Billy was responsible for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 4. Billy got what he deserved because of what he did or said.............1 2 3 4 5 6 7 5. The events of that day were Billy’s fault because of what he did or said.....................1 2 3 4 5 6 7 6. Billy is to blame for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 7. Billy was responsible for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 8. Billy got what he deserved because of the kind of person he is........1 2 3 4 5 6 7 9. The events of that day were Billy's fault because of the kind of person he is................1 2 3 4 5 6 7 10. Billy is to blame for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 137 APPENDIX C (cont'd.) Sandra Shale's behavioral response toward Billy was justified.................1 Sandra Shale's actions toward Billy were appropriate to what Billy did.............................1 Sandra Shale really had no alternatives other than what she did.......1 Most people would have responded as Sandra Shale did........................1 I feel similar to Billy....................1 I identify with Billy......................1 Billy reminds me of myself.................1 I feel a sense of sameness with Billy.................................1 If the average parent had a child who did what Billy did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry ;:::::quite angry g:::::enraged N NNNN N N U UUUU U uh bébfi uh ab 0' 0000! 0 0| 0 0000 0 0 Q QQQQ Q Q 138 APPENDIX C (cont'd.) Rachel, the 8 year old daughter of Judith Mills, was watching television while her Mom was cooking dinner. Rachel thought it would be funny to turn the volume up to the maximum. Judith became very upset by this and repeatedly called for Rachel to turn the volume down. Rachel threw a sofa pillow at her mother and replied with: "Don't bother me!" Finally, Judith hit Rachel repeatedly with the wooden spoon she had in her hand. Rachel's neck became bruised. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 - strongly disagree 4 = neutral 2 = disagree 5 = agree somewhat 3 8 disagree somewhat 6 = agree 7 = strongly agree 1. Rachel's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 2. This event was severe and traumatic for Rachel.......................1 2 3 4 5 6 3. Rachel was responsible for what happened to her because of what she did or said....................1 2 3 4 5 6 4. Rachel got what she deserved because of what she did or said............1 2 3 4 5 6 5. The events of that day were Rachel's fault because of what she did or said....................1 2 3 4 5 6 6. Rachel is to blame for what happened to her because of what she did or said....................1 2 3 4 5 6 7. Rachel was responsible for what happened to her because of the kind of person she is...............1 2 3 4 5 6 8. Rachel got what she deserved because of the kind of person she is.......1 2 3 4 5 6 9. The events of that day were Rachel's fault because of the kind of person she is...............1 2 3 4 5 6 10. Rachel is to blame for what happened to her because of the kind of person she is...............1 2 3 4 5 6 11. 12. 13. 14. 15. 16. 17. 18. 19. 139 APPENDIX C (cont'd.) Judith Nills’ behavioral response toward Rachel was justified................1 Judith Nills' actions toward Rachel were appropriate to what Rachel did............................1 Judith Mills really had no alternatives other than what she did.......1 Most people would have responded as Judith Nills did........................1 I feel similar to Rachel...................1 I identify with Rachel.....................1 Rachel reminds me of myself................l I feel a sense of sameness with Rachel................................1 If the average parent had a child who did what Rachel did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry ;:::::quite angry g:::::enraged N NNNN N N U UUUU U U uh hhbuh uh ch 0‘ 0000' 0| 0| 0 0000 0 0 Q QQQQ Q Q 140 APPENDIX C (cont'd.) Patrick, an 8 year old boy, was given a new Winter coat from his father, William Spence. When he returned from school, Patrick's coat was covered with mud. William asked Patrick what happened and Patrick said he didn't know. His father asked him several times for an explanation. Patrick just stood quietly. Finally, William took off his belt and strapped Patrick several times with it. Patrick later developed bruises. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 = strongly disagree 4 = neutral 2 = disagree 5 = agree somewhat 3 = disagree somewhat 6 8 agree 7 a strongly agree 1. Patrick's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Patrick......................1 2 3 4 5 6 7 3. Patrick was responsible for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 4. Patrick got what he deserved because of what he did or said.............l 2 3 4 5 6 7 5. The events of that day were Patrick's fault because of what he did or said.....................1 2 3 4 5 6 7 6. Patrick is to blame for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 7. Patrick was responsible for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 8. Patrick got what he deserved because of the kind of person he is........1 2 3 4 5 6 7 9. The events of that day were Patrick’s fault because of the kind of person he is................l 2 3 4 5 6 7 10. Patrick is to blame for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 141 APPENDIX C (cont'd.) William Spence's behavioral response toward Patrick was justified...............1 William Spence's actions toward Patrick were appropriate to what PatriCR didOOOOOOOOOOOOOOOOOOOOOOOOOOII William Spence really had no alternatives other than what he did........1 Most people would have responded as William Spence did......................1 I feel similar to Patrick..................1 I identify with Patrick....................l Patrick reminds me of myself...............1 I feel a sense of sameness with Patrick...............................1 If the average parent had a child who did what Patrick did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry ;:::::guite angry g:::::enraged N NNNN N N UUUU U U U .5 «hitch-h uh ab 0! 000101 0 01 0 0000 0 0 Q QQQQ Q Q ' L 142 APPENDIX C (cont'd.) Veronica Weiner was taking out the garbage. Her 8 year old son, Phillip was with her. As she was about to drop in the trash, she noticed the expensive toy she had just purchased for him was sitting broken in the yard. She pointed this out to Phillip and asked him what happened. He just said: "I’m not sure." Veronica asked him again and he stood there quietly. Veronica hit Phillip several times with the trash can lid. His left hand became bruised. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 = strongly disagree 4 a neutral 2 = disagree 5 = agree somewhat 3 = disagree somewhat 6 = agree 7 a strongly agree 1. Phillip's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Phillip......................l 2 3 4 5 6 7 3. Phillip was responsible for what happened to him because of what he did or said.....................l 2 3 4 5 6 7 4. Phillip got what he deserved because of what he did or said.............1 2 3 4 5 6 7 5. The events of that day were Phillip's fault because of what he did or said.....................1 2 3 4 5 6 7 6. Phillip is to blame for what happened to him because of what he did or said.....................1 2 3 4 5 6 7 7. Phillip was responsible for what happened to him because of the kind of person he is................l 2 3 4 5 6 7 8. Phillip got what he deserved because of the kind of person he is........1 2 3 4 5 6 7 9. The events of that day were Phillip's fault because of the kind of person he is................1 2 3 4 5 6 7 10. Phillip is to blame for what happened to him because of the kind of person he is................1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 143 APPENDIX C (cont'd.) Veronica Weiner's behavioral response toward Phillip was justified...............1 Veronica Weiner’s actions toward Phillip were appropriate to what Phillip did...........................1 Veronica Weiner really had no alternatives other than what she did.......1 Most people would have responded as Veronica Weiner did.....................1 I feel similar to Phillip..................1 I identify with Phillip....................1 Phillip reminds me of myself. ....... .......1 I feel a sense of sameness with Phillip...............................1 If the average parent had a child who did what Phillip did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry ;:::::quite angry $:::::....g.. NNNN N N UUUU uh. «huh-bub uh uh N U U 0' 001010 0' U' 0000 0 0 Q QQQQ Q Q 144 APPENDIX C (cont'd.) Jessica, Harold Bateman's 8 year old daughter was playing in her room by herself. Harold opened the door to look inside. He noticed Jessica's toys were scattered all over her room in a mess. Harold asked her why she had not cleaned her room yet. She said she wasn't sure. He asked her again and she didn't say anything. Harold took off his belt, and strapped her repeatedly on the bottom. She subsequently developed bruises. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 = strongly disagree 4 = neutral 2 = disagree 5 = agree somewhat 3 a disagree somewhat 6 = agree 7 = strongly agree 1. Jessica’s behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Jessica......................1 2 3 4 5 6 7 3. Jessica was responsible for what happened to her because of what she did or said....................1 2 3 4 5 6 7 4. Jessica got what she deserved because of what she did or said............1 2 3 4 5 6 7 5. The events of that day were Jessica's fault because of what she did or said....................1 2 3 4 5 6 7 6. Jessica is to blame for what happened to her because of what she did or said....................1 2 3 4 5 6 7 7. Jessica was responsible for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 8. Jessica got what she deserved because of the kind of person she is.......1 2 3 4 5 6 7 9. The events of that day were Jessica's fault because of the kind of person she is...............1 2 3 4 5 6 7 10. Jessica is to blame for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 145 APPENDIX C (cont'd.) Harold Bateman’s behavioral response toward Jessica was justified...............1 Harold Bateman's actions toward Jessica were appropriate to what Jessica did...........................1 Harold Bateman really had no alternatives other than what he did........1 Most people would have responded as Harold Bateman did......................1 I feel similar to Jessica..................1 I identify with Jessica....................1 Jessica reminds me of myself...............1 I feel a sense of sameness with Jessica...............................1 If the average parent had a child who did what Jessica did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry g:::::quite angry $:::::enraged ””1010 N N UUUU U :5 «bah-bub ah ah N U U! 0000' 0' 01 0 0000 0 0 Q QQQQ Q Q \ 146 APPENDIX C (cont'd.) Catherine is Helen Willis' 8 year old daughter. Helen noticed her pearl earrings were missing from her jewelry box again. Helen walked into Catherine's room, and noticed the earrings sitting on the dresser. She asked Catherine why she had taken her earrings. Catherine said that she didn't know. Helen asked her again. When Catherine just looked away quietly, Helen hit her repeatedly with a book. Catherine's left thigh later became bruised. Please indicate the extent to which you agree or disagree with each of the following statements using the following scale: 1 - strongly disagree 4 = neutral 2 = disagree 5 2 agree somewhat 3 2 disagree somewhat 6 = agree 7 = strongly agree 1. Catherine's behavior was provocative (e.g., mischievous, insolent, or aggressive)......1 2 3 4 5 6 7 2. This event was severe and traumatic for Catherine....................1 2 3 4 5 6 7 3. Catherine was responsible for what happened to her because of what she did or said....................1 2 3 4 5 6 7 4. Catherine got what she deserved because of what she did or said............1 2 3 4 5 6 7 5. The events of that day were Catherine's fault because of what she did or said....................1 2 3 4 5 6 7 6. Catherine is to blame for what happened to her because of what she did or said....................1 2 3 4 5 6 7 7. Catherine was responsible for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 8. Catherine got what she deserved because of the kind of person she is.......1 2 3 4 5 6 7 9. The events of that day were Catherine's fault because of the kind of person she is...............1 2 3 4 5 6 7 10. Catherine is to blame for what happened to her because of the kind of person she is...............1 2 3 4 5 6 7 11. 12. 13. 14. 15. 16. 17. 18. 19. 147 APPENDIX C (cont'd.) Helen Willis' behavioral response toward Catherine was justified.............1 Helen Willis' actions toward Catherine were appropriate to what Catherine did.........................1 Helen Willis really had no alternatives other than what she did.......1 Most people would have responded as Helen Willis did........................1 I feel similar to Catherine................1 I identify with Catherine..................1 Catherine reminds me of myself.............1 I feel a sense of sameness with Catherine.............................1 If the average parent had a child who did what Catherine did, how angry would the parent be? 1.....mildly irritated §:::::somewhat angry ;:.:::quite angry $.::::enraged NNNN N N UUUU U N U h chub-bub uh :9- UI mount» 0' 0| 0 0000 0 0 Q QQQQ Q Q APPENDIX D DUNCAN SOCIO-ECONOMIC INDEX APPENDIX D DUNCAN SOCIO-ECONOMIC INDEX Think back to when you were about 11_y§az§_91d. 1A. What kind of work was your father doing? (For example: electrical engineer, stock clerk, farmer.) 18. What were his most important activities or duties? (For example: kept account books, filed, sold cars, operated printing press, finished concrete.) 1C. What kind of business or industry was this? (For example: TV and radio mfg., retail shoe store, State Labor Dept., farm.) 1D. Was he: (Circle one below). A) an employee of a PRIVATE company, business or individual for wages, salary, or commissions?..........PR B) a GOVERNMENT employee (federal, state, county, or local government)?..........................GOV C) self-employed in OWN business, professional practice, or farm? own business not incorporated (or farm)......OWN own business incorporated....................INC D) working WITHOUT PAY in a family business or farm?......................................WP 149 150 APPENDIX D (cont'd.) Think back to when you were about 11_ygars_gld. 1A. What kind of work was your mother doing? (For example: electrical engineer, stock clerk, farmer.) 18. What were her most important activities or duties? (For example: kept account books, filed, sold cars, operated printing press, finished concrete.) 1C. What kind of business or industry was this? (For example: TV and radio mfg., retail shoe store, State Labor Dept., farm.) 10. Was she: (Circle one below). A) an employee of a PRIVATE company, business or individual for wages, salary, or commissions?..........PR B) a GOVERNMENT employee (federal, state, county, or local government)?..........................GOV C) self-employed in OWN business, professional practice, or farm? own business not incorporated (or farm)......OWN own business incorporated....................INC D) working WITHOUT PAY in a family business or farm?......................................WP (2) Are you: male or female? How old are you? (3) Are you: a) Black b) White c) Oriental d) Hispanic e) Other (4) Are you: a) Jewish b) Catholic c) Protestant d) Atheist e) Other (5) Do you have any children of your own: yes or no? If so, how many? (6) How many brothers do you have? (7) How many sisters do you have?_____ (8) Have you ever been in psychotherapy or counseling: yes or no?_____ (9) If so, approximately how many sessions did you have? APPENDIX E INSTRUCTIONS/LETTER OF CONSENT APPENDIX E INSTRUCTIONS/LETTER OF CONSENT Before you start, tear of, and keep the cover sheet, which has your ID number on it (the previous page). The following is a study of family environment and your personal attitudes. Our purpose is to see how you feel about interpersonally conflicting family situations. We will be using only the questionnaires you have here, and we are surveying your personal attitudes. There are no right or wrong answers. Whatever you honestly feel or believe is what we would like to know. Your identity will be kept completely anonymous. No one (including the experimenter) will know who completed which questionnaires. Please try to be as honest as possible. The first testing session last approximately 1.5 hours. If you choose to take part in a second administration, that will last a half hour. Your participation in this study is voluntary. Participation may be terminated at any time. Should you feel upset by some of the things you read, you may terminate participation. If you feel you understand the survey, and what is involved, then begin. You indicate your voluntary consent agreement to participate by completing and returning this questionnaire. My name is Rob Muller (351-4561). Should you have any concerns regarding this study, don't hesitate to call. 152 APPENDIX E INSTRUCTIONS FOR SECOND ADMINISTRATION APPENDIX F INSTRUCTIONS FOR SECOND ADMINISTRATION Thank you for returning for a second administration. Just like last time, please read each scenario carefully and answer the questions honestly. Your identity is kept completely anonymous. This session will last about a half hour. Many of the questions will be questions we have already asked you. Answer them in terms of how you honestly feel It is essential that you write down the ID number we gave you last time on this questionnaire. You may begin. For any questions regarding this study, don't hesitate to contact: Robert Muller Department of Psychology Michigan State University 351-4561 154 APPENDIX G FEEDBACK FORM APPENDIX G FEEDBACK FORM W The following is a brief explanation of the main ideas behind the study in which you participated. Many researchers have discovered that in Western culture, there is a tendency to see victims of unfortunate life events as responsible for their misfortune. Thus, rape victims are often (incorrectly) perceived as responsible. This phenomenon has been referred to as "blaming the victim." The present study investigates the tendency to see abused children as responsible for the punitive behaviors of their parents. Various researchers have found that abused children tend to blame themselves for their parents' harsh punitive treatment. It was hypothesized that individuals who were physically abused as children, blamed themselves. Having blamed themselves, these persons developed the attitude that children are responsible for their parents' harsh punitiveness. A competing hypothesis to the one aforementioned, is that adult survivors of abuse learned early on that the world can be very unfair. As such, they would believe that abused children are not to blame, and that these children are merely the recipients of unfair treatment. Several researchers have discovered that provocative children are abused more often than nonprovocative ones. In this investigation, it was proposed that provocative children will be seen as more blameworthy than nonprovocative children. Based on prior research, we also hypothesized that male children will be perceived as more blameworthy than female children, and that male subjects will victim blame more than female subjects will. If you have any questions regarding this study, please call: Robert Muller 351-4561 Thank you for participating in this investigation. 156 LIST OF REFERENCES LIST OF REFERENCES Alexander, C. S. (1980). The responsible victim: Nurses' perceptions of victims of rape. Jon:nal_gf_fleal;n and_Sesial_Bahayior. 21. 22-33- Allen, D. M., 8 Tarnowski, K. J. (1989). Depressive characteristics of physically abused children. , ll, 1-11. Amsterdam, 8., Brill, M., Bell, N. W. 8 Edwards, D. (1979). Coping with abuse: Adolescents’ views. , 278-284. Anderson, S. C., & Lauderdale, M. L. (1982). Characteristics of abusive parents: A look at self- esteem. Child_Ahu§e_and_nglect. 6. 285-293- Aronoff, J., 8 Wilson, J. P. (1985). 2g:sgnali§y_ln_thg social_prggg§§. Hillsdale, N.J.: Lawrence Erlbaum. Azar, S. T., & Wolfe, D. A. (1989). Child abuse and neglect. In E. J. Mash & R. A. Barkley (Eds.), Treatment_ef_childh92d_disgrders (pp- 451-489)- New York, N.Y.: Guilford Press. Bakan. D. (1971). Slanghter_ef_ths_innecents- San Francisco: Jossey-Bass, Inc. Baker, H. 3., & Baker, M. N. (1987). Heinz Kohut's self Psychology: An overview. American_fleurnal_of I 1.4.5. 1-9- Barahal, R. M., Waterman, J., 8 Martin, H. P. (1981). The social cognitive development of abused children. 508-516. Berger, A. M., Knutson, J. F., Mehm, J. G., & Perkins, K. A. (1988). The self-report of punitive childhood experiences of young adults and adolescents. , 12, 251-262. Briere, J., & Runtz, M. (1990). Differential adult symptomatology associated with three types of child abuse histories. Child_Ahuse_and_Nsslsst. 11. 357-364. 158 159 Burt, M. (1980). Cultural myths and support for rape. . 35. 217-230. Calhoun, L. G., Selby, J. W., Cann, A. 8 Keller, G. T. (1978). The effects of victim physical attractiveness and sex of respondent on social reactions to victims of rape. , 191-192. Campbell, R. T. 8 Parker, R. N. (1983). Substantive and statistical considerations in the interpretation of multiple measures of SES. Sgglal_£gzcgsl_§z, 450-466. Carroll, J. C. (1977). The intergenerational transmission of family violence: The long-term effects of aggressive behavior. Aggressiye_8ehaxieri_3. 289- 299. Christensen. L- B- (1985). Exnerimental_methodology- Toronto: Allyn 8 Bacon. Cohen, J. (1978). Partialled products are interactions: partialled powers are curve components. Psychologisal_fiulletin. 85. 858-866- Cohen, L. H., McGowan, J., Fooskas, S., 8 Rose, S. (1984). Positive life events and social support and the relationship between life stress and psychological disorder. 2§¥£thQ§¥o 12. 557’537- Cohen, P. M. (1984). Violence in the family: An act of loyalty? Eeyshgtheranyi_21. 249-253. Coller, S. A., 8 Resick, P. A. (1987). WOmen's attributions of responsibility for data rape: The influence of empathy and sex-role stereotyping. , 115-125. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrikal_lfi, 297-334. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. 85. Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. and_Sesia1.2522helogyi_44.113- 126. 160 Deitz, s. R., Blackwell, K. T., Daley, P. C., 8 Bentley, B. J. (1982). Measurement of empathy toward rape victims and rapists. Journal_ef_zersgnality_and Social_zaychelosxi_13. 372-384. Deitz, s. R., Littman, M., 8 Bentley, B. J. (1984). Attribution of responsibility for rape: The influence of observer empathy, victim resistance, and victim attractiveness. s§x_Bngsl_lQ, 261- 279. Dymond, R. F. (1949). A scale for the measurement of empathic ability. ll, 127-133. Endler, N. S. (1983). Interactionism: A personality model, but not yet a theory. In M. M. Page (Ed.), Personalitx_:current_theory_and_research (pp. 155- 200). Lincoln, Nebraska: University of Nebraska Press. Engfer, A. 8 Schneewind, K. A. (1982). Causes and consequences of harsh parental punishment: An empirical investigation in a representative sample of 570 German families. Child_abuse_and_neglecti 6, 129-139. Evans, M. G. (1991). The problem of analyzing multiplicative composites: Interactions revisited. American_zsychelogist. 16. 6-15. Fleming, R., 8 Baum, A. (1986). Social support and stress: The buffering effects of friendship. In V. J. Derlaga, 8 B. A. Winstead (Eds.), £rlgndship_and social_interactien (pp. 207-225). New York: Springer-Verlag. Fleming, R., Baum, A., Gisriel, M. M., 8 Gatchel, R. J. (1982). Mediating influences of social support at Three Mile Island. Journal_9f_numan_5tress. a. 14-22. Funder, D. C. (1991). Global traits: A neo—Allportian approach to personality. 2529hglgglg§l_§gigng§, 2, 31-39. Gaudin, J. M., 8 Pollane, L. (1983). Social networks, streSS. and child abuse. Childran_xeuth_seryices 3:11.31. 5.: 91‘102 . 161 Gelles, R. J. (1976). Demythologizing child abuse. Eemlly Coordinator. 2:, 135-141. Ghiselli, E. E., Campbell, J. P., 8 Zedeck, S. (1981). San Francisco: W. H. Freeman and Co. Gil, D. (1970). ' ' . Cambridge, Mass.: Harvard University Press. Gillespie, D. F., Seaberg, J. R. 8 Berlin, S. (1977). Observed causes of child abuse. yieelnelegyl_z, 342- 349 0 Gold, E. R. (1986). Long-term effects of sexual victimization in childhood: An attributional approach. ' , 51, 471-475. Green, A. H. (1982). Child abuse. In J. R. Lachenmeyer 8 M. 8- Gibbs (Eds.). EsYohooafhology_in_ohildhood. New York: Plenum. Green, S. B., Lissitz, R. W., 8 Mulaik, S. A. (1977). Limitations of coefficient alpha as an index of test unidimensionality. Ednee;lenel_end Psycholooioal_noasuromonti_31. 827-838. Hauser, R. M. 8 Featherman, D. L. (1977). Ihe_pzeeeee_ef stratifioationi__Tronds_and_analyoos. New York: Academic Press. Helfer, R. E. (1987). The developmental basis of child abuse and neglect: An epidemiological approach. In R. E. Helfer 8 R. s. Kempe (Eds.), The (pp. 60-80). Chicago: University of Chicago Press. Helfer, R. E., McKinney, J., 8 Kempe, R. (1976). Arresting or freezing the developmental process. In R. E. Helfer 8 C. H. Kempe (Eds.), thld_ebn§e_and Cambridge, Mass.: Ballinger. Herrenkohl, R. C., Herrenkohl, E. C. 8 Egolf, B. P. (1983). Circumstances surrounding the occurrence of child maltreatment. PsycholooYi_ol. 424-431. 162 Herzberger, S. D., Potts, D. A. 8 Dillon, M. (1981). Abusive and nonabusive parental treatment from the child's perspective. WM. 81-90. Herzberger, S. D. 8 Tennen, H. (1985). "Snips and snails and puppy dog tails": Gender of agent. Sex Bolofii_12. 853-865. Hjorth, C. W. 8 Ostrov, E. (1982). The self-image of physically abused adolescents. 1enrnel_ef_xenth anLAdolosoenomJl. 71-76. Hoffman, M. L. (1977). Empathy, its development and prosocial implications. In C. B. Keasey (Ed.), ' (Vol 25). Lincoln: University of Nebraska Press. Hogan, R. (1969). Development of an empathy scale. , 307- 316. Hunter, J. E. (1986, October). . Workshop monograph presented at University of Iowa. Iowa city, Iowa. Hunter, J. E., 8 Cohen, S. H. (1969). PACKAGE: A system of computer routines for the analysis of correlational data. Moasnromon11_22. 697-700- Hunter, J. E. 8 Gerbing, D. W. (1982). Unidimensional measurement, second order factor analysis, and causal models. In B. M. Straw 8 L. L. Cummings (Ede-L W (Vol. 4). Greenwich, Conn.: JAI Press. Hunter, J. E., Gerbing, D. W., 8 Boster, F. J. (1982). Machiavellian beliefs and personality: Construct invalidity of the Machiavellianism dimension. 1293-1305. Hunter, J. E., Gerbing, D. W., Cohen, S. H., 8 Nicol, T. S. (1980). EAQKAEE_12fiQI__A_§Y§§£m_o£_EQBTEAN (Available from Academic Computing Services, Baylor University, Waco, Texas.) 163 Isaacs, C. (1981). A brief review of the characteristics of abuse-prone parents. WW. 5- 8. Janoff-Bulman, R. (1979). Characterological versus behavioral self-blame: Inquiries into depression and rape. 2§¥§h919321_12. 1798-1809- Jenkins, M. J. 8 Dambrot, F. M. (1987). The attribution of date rape: Observers' attitudes and sexual experiences and the dating situation. Jenznel_ef W11. 875-895. Jensen, I. W., 8 Gutek, B. A. (1982). Attributions and assignment of responsibility in sexual harassment. W. 35. 121-136. Joreskog, K. G. (1978). Structural analysis of covariance and correlation matrices. Peyenemetrikel_11, 444- 447. Kanekar, S. 8 Kolsawalla, M. B. (1980). Responsibility of a rape victim in relation to her respectability, attractiveness and provocativeness. Jen:nel_ef SEW. 153-154- Kaufman, J. 8 Zigler, E. (1987). Do abused children become abusive parents? W141: orthoooyohiatzyi_§z. 186-192. Kazdin. A- E. (1980). new peyenelegy. New York: Harper 8 Row. Kempe, R. S. 8 Kempe, C. H. (1976). Assessing family pathology. In R. E. Helfer 8 C. H. Kempe (Eds. ), Child . Cambridge, Mass.: Ballinger. Xempe, R. S. 8 Kempe, C. H. (1978). ghlld_ebnee. Cambridge, Mass.: Harvard University Press. Kerr, N. L. 8 Kurtz, S. T. (1977). Effects of a victim's suffering and respectability on mock juror judgments: Further evidence on the just world theory. E§¥Qh919911_§. 42-56- Kerr, W. A. 8 Speroff, B. G. (1954). Validation and evaluation of the empathy test. Jenrnel_ef 92W. 369-376- 164 Koestner, R., Franz, C., 8 Weinberger, J. (1990). The family origins of empathic concern: A 26-year longitudinal study. Journal_of_zorsonality_and_5ooial W: 709‘7170 Krahe, B. (1988). Victim and observer characteristics as determinants of responsibility attributions to victims of rape. Joornal_of_Aooliod_Sooial_Bsyoholooyi_18. 50- 58. Krulewitz, J. E. 8 Nash, J. E. (1979). Effects of rape victim resistance, assault outcome, and sex of observer on attributions about rape. Journal_of_zorsonalityi 51, 557-574. Larrance, D. T. 8 Twentyman, C. T. (1983). Maternal attributions and child abuse. Jeurnel_ef_Abnermal Eoyoholooxi_22. 449-457- Ledray, L. E. (1986). Reeeyerlng_frem_reee. New York: Holt. Lerner. M. J. (1980). Tho_boliof_in_a_iust_oorld. New York: Plenum. Levenson, H. (1981). Differentiating among internality, powerful others, and chance. In H. M. Lefcourt (Ed.). Bosoaroh_aith_tho_loous_of_oontrol oonstrnoti__Assossmont_mothods - . New York: Academic Press. Libow, J. A. 8 Doty, D. W. (1979). An exploratory approach to self-blame and self-derogation by rape victims. Amorioan_2oornal_of_orthoooyohiotrxi_52. 670-679. Lieh-Mak, F., Chung, S. Y., 8 Liu, Y. W. (1983). Characteristics of child battering in Hong Kong: A controlled study. Britleh_lenrnel_ef Esyohiatryi_llz. 89-94. Ma, L., 8 Smith, J. B. (1985). Individual and social correlates of the just world belief: A study of Taiwanese college students. Eeyehelegleel BQDQISS. 51. 35'33- MacLean, M. J., 8 Chown, S. M. (1988). Just world beliefs and attitudes toward helping elderly people: A comparison of British and Canadian university students. Human_noyoloomont. 25. 249-259- 165 Maden, M. F. 8 Wrench, D. F. (1977). Significant findings in child abuse research. yleelmelegyl_z, 196-224. Main, M., 8 George, C. (1985). Responses of abused and disadvantaged toddlers to distress in agemates: A study in the day care setting. neyelepneneel Bfi¥£h919§¥1_21. 407-412- McCrae, R. R., 8 Costa, P. C., Jr. (1984). Energing_llyee, W. Boston: Little. Brown. Mehrabian, A. 8 Epstein, N. (1972). A measure of emotional empathy. W. 525-543 - Meyer, C. 8 Taylor, S. E. (1986). Adjustment to rape. . 50. 1225- 1234. Miller, K. A. 8 Miller, E. K. (1983). Self-reported incidence of physical violence in college students . loumaLoLthLAmorioanJollogLof W. 63-65 . Miller, P. A., 8 Eisenberg, N. (1988). The relation of empathy to aggressive and externalizing/antisocial behavior. W. 324-344. Motta, R. W., 8 Tiegerman, S. (1979). Perceptions of justice: An adolescent view. Jenznel_ef WW. 2. 151-157- Ney, P. G., Moore, C., McPhee, J., 8 Trought, P. (1986). Child abuse: A study of the child’s perspective. MW. 511-518. O'Quin, K., 8 Vogler, C. C. (1990). Use of the just world scale with prison inmates: A methodological note. , ZQ, 395-400. Paulsen, K. A. (1979). Attribution of fault to a rape victim as a function of locus of control. Jeannel_ef W. 131-132- Pelton, L. H. (1978). Child abuse and neglect: The myth of classlessness . Americanioornalm W. 608-617 - Phares, E. J. 8 Lamiell, J. T. (1975). Internal-external control, interpersonal judgments of others in need, and attribution of responsibility. {gunnel mm. 23-28- 166 Phares, E. J. 8 Wilson, K. G. (1972). Responsibility attribution: Role of outcome severity, situational ambiguity, and internal-external control. Jeannel_efi Bensonalityi_lo. 392-406. Rausch, K., 8 Knutson, J. F. (1991). The self-report of personal punitive childhood experiences and those of siblings. Child_Abuso_and_Nooloot. 15. 29-36. Reid, J. B., Patterson, G. R. 8 Loeber, R. (1981). The abused child: victim, instigator, or innocent Rosenberg, M. S. 8 Reppucci, N. D. (1983). Abusive mothers: Perceptions of their children's behavior. E§¥§h919911_§1. 674-632- Rouse, L. P. (1984L Models, self-esteem, and locus of control as factors contributing to spouse abuse. Yiotimolooy. 2. 130- -l41. Rubin, z. 8 Peplau, A. (1973). Belief in a just world and reactions to another's lot: A study of participants in the national draft lottery. 22, 73-93. Russell, G. M., 8 Jorgenson, D. O. (1978). Religious group membership, locus of control, and dogmatism. Psychological_3ooorta. 12. 1099-1102- Ryan, W. (1971). Blenlng_;ne_yle;lm. New York: Random House. Sarason, I. G., Sarason, B. R., Potter, E. H., 8 Antoni, M. H. (1985). Life events, social support, and illness. Eoyohosomatio_Nodioino. 41. 156-163. Schoeneman, T. J., Hollis, J. F., Stevens, V. J., 8 Fischer, K. (1988). Recovering stride versus letting it slide: Attributions for ”slips" following smoking cessation treatment. ESYoholooY_ond_Hoolth. 2. 335-347. Shaver, K. G. (1970). Defensive attribution: Effects of severity and relevance on the responsibility assigned for an accident. iflfi.§9§ifll.£§¥§hfll§§¥ lip 101- 113. 167 (1986). On causality, Shaver, K. G., 8 Drown, D. A theoretical responsibility, and self-blame: note. Psychology. 59. 697-702. (1979). Child abuse and deprivation: Soul Shengold, L. L. murder. Association. 21. 533-559. Simons R. L., Whitbeck, L. B., Conger, R. D., 8 Chyi-In, W. Intergenerational transmission of harsh (i991). parenting. W. 2.2. 159-171. Skrypnek. 3- J. (1980). MW Unpublished doctoral dissertation, University of Minnesota, Minneapolis, Minnesota. ' A Smith, J. E. (1984). Non-accidental injury to children. review of behavioral interventions, Beheyler Bosoaroh_ond_Thorooyi_22. 331-347- Smith, S. L. (1984). Significant research findings in the Journal_of_oontomoorary etiology of child abuse. SooiaI_Hork. 55. 337-346. Smith R. E., Keating, J. P., Hester, R. K. 8 Mitchell, H. E. (1976). Role and justice considerations in the ' Journal_of;Bosoaroh_in attribution of responsibility. Personalityi_lo. 346-357- (1974). Internal-external control and the Sosis, R. H. perception of responsibility of another for an accident. Journal_of_zoraonality_ano_Social EEXQDQIQ§¥+_IQ. 393-399- ' A Steele, B. (1975). Working with abusive parents. psychiatrist's view. ghlld:en_Tedeyl_e, 3. Steele, B. F. (1986). Notes on the lasting effects of early child abuse throughout the life cycle. Child_Aboso_and_Nogloot. lo. 283-291. Stollak G. E., Scholom, A., Kallman, J., 8 Saturansky, C. Insensitivity to children: Responses of (1973). undergraduates to children in problem situations. Journal_of_Abnormal_ohilo_zsyoholooyi_2.169-180. Stotland S., Mathews, K., Sherman, S., Hansson, R., 8 Richardson, B. (1978). helping. Beverly Hills, Calif.: Sage. 168 Straker, G. 8 Jacobson, R. S. (1981). Aggression, emotional maladjustment, and empathy in the abused child. noooloomontal_Esyoholooyi_11. 762-765. Straus, M. A. (1989). Sealee. Durham: Family Research Laboratory. Straus, M. A. 8 Gelles, R. J. (1986). Societal change and change in family violence from 1975 to 1985 as revealed by two national surveys. , 465-479. Straus, M. A. 8 Gelles, R. J. (1989). £hyeleel_glelenee_in Amorioan_familioa1__Eisk_faotors_ono_aoaotations . New Brunswick, NJ: Transaction Books. Suedfeld, P., Hakstian, A. R., Rank, D. S., 8 Ballard, E. J. (1985). Jonrnal_of_Aooliod_Sooial_2oyoholooy. 15. 285-311. Symonds, M. (1975). Victims of violence: Psychological effects and aftereffects. The_Amerieeh_Jenrnal_ef Psychoanalysisi_3i. 19-26. Thornton, B. (1984). Defensive attribution of responsibility: Evidence for an arousal-based motivational bias. 2orsonality_and.§ooial.25¥oholooxi_lo. 721- -734. Thornton, B., Robbins, M. A., 8 Johnson, J. A. (1981). Social perception of the rape victim's culpability: The influence of respondents' personal-environmental causal attribution tendencies. Hunen_3eleglehe, 11, 225-237. Thornton, B., Ryckman, R. M., 8 Robbins, M. A. (1982). The relationships of observer characteristics to beliefs in the causal responsibility of victims of sexual assault. finnen_3ele;lehe, 15, 321-330. Toro, P. A. (1982). Developmental effects of child abuse: A review. child_abo5o_and_noolooti_o. 423-431. Trickett, P. K., Aber, J. L., Carlson, V., 8 Cicchetti, D. (1991). Relationship of socioeconomic status to the etiology and developmental sequelae of Physical child abuse. Dexoloonontal_zsyoholoox. 22, 148-158. 169 Tunnell, G. B. (1977). Three dimensions of naturalness: An expanded definition of field research. Psycholooioa1_nulletin. aa. 426-437. Vondra, J. I., Barnett, D., 8 Cicchetti, D. (1990). Self- concept, motivation, and competence among preschoolers from maltreating and comparison families. Child_Abuse_and_Neoleot. 11. 525-540. Ward, C. (1988). The attitudes toward rape victims scale: Construction, validation and cross-cultural applicability. 2syono1oox_of_Honen_onarterlyi_12. Waterman, C. K. 8 Foss-Goodman, D. (1984). Child molesting: Variables relating to attribution of fault to victims, offenders, and nonparticipating Parents. Jonrnal_of_Sex_Researohi_zo. 329-349. Webster-Stratton, C. (1985). Comparison of abusive and nonabusive families with conduct-disordered children. Amerioan_lournal_of_ortnoosyohiatryi 55, 59-69. Wilcox, B. L. (1981). Social support, life stress, and psychological adjustment: A test of the buffering hypothesis. Amerioan_Journal_of_oommnnity EEXQDQIQQX. 2. 371-386- Witt, L. A. (1988). Urban-nonurban differences in social cognition: Locus of control and perceptions of a just world. Tne_Journa1_of_Sooia1_2syono1ooy. l22, 715-717. Wolfe, D. A. (1985). Child-abusive parents: An empirical review and analysis. EsyonoIooioa1_anlletin. 21. 462-482. Zuckerman, M., 8 Gerbasi, K. C. (1977). Dimensions of the I-E scale and their relationship to other personality measures. Edneetienel_and Esyoholooioal_neasnrement. 31. 159-175- Zuckerman, M., Gerbasi, K. C., 8 Marion, S. P. (1977). Correlates of the just world factor of Rotter's I- E scale. Ueasurement. 31. 375-381- HICHIGRN STATE UNIV. LIBRRRIES ||HIlIllllIVHIIHIIIWIIWll!IllllmlNHIIHDIIIWHI 31293009116918