RETURNING MATERIALS: MSU Place in book drop to ‘b-LIBRARIES remove this checkout from '-—_ your record. FINES will be charged if book is returned after the date stamped below. 9w "mi?” ‘ 9 .. 55M?” ,A,».__. .... .o- -1. ’ _‘._'w($< 3M.’ flu A AN ANALYSIS OF DIFFERENCES BETWEEN BATTERED AND NON-BATTERED WOMEN WITH RESPECT TO SEX ROLE ACCEPTANCE, LIFE HISTORIES. AND PERSONAL ADJUSTMENT By Maxine A. Thome A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1982 © 1983 MAXINE ALYCE THOME All Rights Reserved AN ANALYSIS OF DIFFERENCES BETWEEN BATTERED AND NON-BATTERED WOMEN WITH RESPECT TO SEX ROLE ACCEPTANCE, LIFE HISTORIES, AND PERSONAL ADJUSTMENT by Maxine A. Thome This investigation attempted to isolate psychological differences between battered and non—battered women. The research questions and hypotheses fit into three areas: (1) Sex Role Acceptance; (2) Life Histories, and (3) Personal Adjustment. Failure to find a suitable questionnaire led the investigator to design her own. This questionnaire was in forced choice format, and included scales devised by this investigator, the Bem Sex Role Inventory, the Dyadic Adjustment Scale, the Parental Attitudes and Attributes scales, and the Personal Respect subscale from the Internal Versus External Control of Reinforcement Scale. The design used was the Static-Group Comparison in which women who had been beaten by their mates were compared with a similar group of non-battered women.~ Battered women from the Washtenaw County shelter for battered women, and non-battered women from self- enrichment courses offered through the lngham County YWCA volunteered to fill out the questionnaire. The groups were similar in that they were both considered to be doing something for themselves, and were of similar marital status, income, age, and educational level. The null hypothesis, stating that there would be no difference between groups, was tested by a MANOVA and when a general area Maxine A. Thome was found to be significant (p < .01) a univariate analysis of variance was done to determine which variables were significant. In addition, frequency counts were completed on several one-item scales. The results indicated failure to reject the null hypothesis for Section 1: Sex Role Acceptance, and Section 2: Life Histories. There are both theoretical and methodological reasons for this failure. Section 3: Personal Adjustment, was significant (p = .001) and the null hypothesis was rejected and may indicate that a major cause of wife battering is the method of communication between the batterer and his mate. There is support for the idea that the battered woman feels competent with adequate self-esteem outside the home, but that her feelings of competency, independence, and self—esteem dissipate when interacting with her mate. Methodological problems that may have influenced the results include the length of the questionnaire, the vocabulary used, and the emotional state of the battered women while living at the shelter. Additionally, the small sample size may have influenced the power and several of the subscales had low reliability. Suggestions for possible remedies to these problems were given. DEDICATION - To Imogen Bowers, my mentor and role model, who through her caring gave me the greatest gift of all -— my self. ACKNOWLEDGMENTS I wish to express my deepest appreciation for the help given me by two of the members of my doctoral committee: Martha Aldenbrand and Verda Scheifley. Although at times my frustration with rewrites blinded me to the value of the changes, I am aware that without Marty's persistence and coaching I would never have completed this research. Marty also taught me that there is life after a literature review. She has been both a mentor and a valued friend. Verda, through her gentle nudges, through her ability to make light of my computer phobia, and her skill in communicating with me as a statistical black hole, also was a major factor in my completion of this project. She was extremely supportive during my many anxiety attacks. I also want to thank Bill Hinds for his support during my final crisis and during the last two months before my orals; and John Schneider for his intellectual and emotional support as well as for his ability to ask questions that led me to look at my data in creative ways. This period in my life would have been much lonelier had it not been for my friendship with Christy Minning. Christy and i started the Counseling Psychology program together and finished our orals within hours of each other. Our friendship grew over the years and provided important caring and support in the final frustrating and painful months and moments before orals. Also, in those near psychotic interludes Christy provided an important validation for my sanity and insantiy. Additionally, I want to thank Susan Wittcoff for her gentle kicks which helped motivate me to keep on plugging away. We became house- mates during the last six months of my working on my dissertation--she moved in when most people want to move out on a doctoral candidate. This friendship is an important one to me. I also want to thank David Rovner, my physician and friend, for his gentle "fatherly" pushes in the last two months. Jo Grandstaff has been incredible and gone far beyond the call of duty as a typist—-she guided me through all steps, including deadlines. I am certain that without Jo's support and guidance I'd be lost in the spaghetti of deadlines and red tape. I also want to thank all of my friends and the women in my women's group who have put up with me when my disposition was all but pleasant. Also, I want to thank the staff and women at the Washtenaw County Shelter and the lngham County YWCA for spending time and energy completing the questionnaire-—without their help this study would never have been completed. Last, but certainly not least, I want to thank my parents who have been supportive of me for the last thirty-two years--after all, they made this whole thing possible. iv TABLE OF CONTENTS Page LIST OF TABLES .............................................. vii LIST OF FIGURES .............................................. viii Chapter I. INTRODUCTION ....................................... 1 The Need for the Study ................................ 1 Purpose ................................................ 3 Research Questions and Hypotheses .................... II II. REVIEW OF LITERATURE .............................. 7 Sex Role Socialization ................................... 8 Learned Helplessness .................................. 12 Morbid Dependency .................................... 13 Isolation ............................................... 14 Violence in Families of Origin .......................... 15 The Three Phase Cycle of Violence ..................... 17 Summary .......................... ' ..................... 19 III. THE QUESTIONNAIRE ................................. 20 History of the Questionnaire .......................... 20 General Description of Content ......................... 22 Item Development ...................................... 24 Definition of Variables ................................. 33 IV DESIGN AND ANALYSIS ............................... 145 Overview ............................................. '15 Population ............................................ 45 Sample ............................................... £16 Procedure ............................................ 118 Testable Hypotheses .................................. 119 Analysis ............................................... 55 Chapter Page V. RESULTS ............................................. 57 Section 1: Sex Role Acceptance ...................... 58 General Hypothesis ................................. 58 Section 2: Life Histories ............................. 59 General Hypothesis .................................. 59 Section 3: Personal Adjustment ...................... 62 General Hypothesis ................................. 62 Conclusion .......................................... 64 Conclusion ............................................ 69 Summary .............................................. 69 VI. DISCUSSION .......................................... 71 General Methodological Problems with the Questionnaire. 71 Discussion of Hypothetical Variables ................... 73 Sex Role Acceptance ................................ 74 Life Histories ...................................... 76 Personal Adjustment ................................ 78 Implications for Future Research .................... 85 REFERENCES ................................................. 87 APPENDICES .................................................. 91 A. QUESTIONNAIRE ..................................... 92 B. DISCUSSION OF TRENDS IN THE DATA FOR SECTION 2: LIFE HISTORIES ....................... 121 General Hypothesis .................................... 121 Findings .............................................. 121 vi Table 3.1 5.7 8.1 LIST OF TABLES Scale Names, Hypotheses Numbers, Item Numbers, and Reliabilities ........................................ Summary of the Characteristics of the Sample: Battered Women (BW) and Non-Battered Women (N-BW) .......... Multivariate Analysis of Variance-~Section 1: Sex Role Acceptance ........................................ Calculation of Group Androgyny Scores for Battered Women (BW), and Non-Battered Women (N-BW) .......... Multivariate Analysis of Variance--Section 2: Life Histories .............................................. Multivariate Analysis of Variance--Section 3: Personal Adjustment ............................................ Frequency Counts--Question 118 ....................... Frequency Counts--Question 121 ....................... Frequency Counts: Question 167 ....................... Univariate Analysis of Variance--Section 2: Life Histories .............................................. vii Page 30 117 58 60 61 63 66 67 68 123 LIST OF FIGURES Figure Page 6.1 Differences on Variables within Section 3: Personal Adjustment .......................................... 80 viii CHAPTER I INTRODUCTION The Need for the Study Although the problem of battered women has been given much attention in the past several years, only limited research has been done to distinguish battered from nonbattered women. Nor has much attention been given to identifying the personality characteristics of battered women. However, battering is a problem of great magnitude, as shown in the following report: 1. In Chicago, a police survey conducted between September 1965 and March 1966 demonstrated that 46.1 percent of all major crimes perpetrated against women, except murder, took place at home (Martin,‘1977, p. 12). 2. In Oakland, California, in 1970, police reportedly responded to more than 16,000 family-disturbance calls in a six-month period (Martin, 1977, p. 12). 3. In Detroit, l1,900 wife-assault complaints were filed in 1972 (Martin, 1977, p. 12). ll. In Washtenaw County, Michigan, the police reported daily family-assault calls in a town of 6,000 (Martin, 1977, p. 13). 5. Nationwide in 1973, according to the FBI, one-fourth of all murders occurred within the family, and one-half of those were husband-wife killings (Martin, 1977, p. 15). 1 6. In lngham County, Michigan, professionals and community agencies in 1976 reported contact with 1,3811 battered women, 511 percent of them battered by their husbands. Of the remaining I~16 percent, a sizable proportion were beaten by men from whom they were legally separated or divorced (Brainard, 1978). 7. Unpublished statistics for the State of Michigan from a reported by the State's Domestic Violence Prevention and Treatment Board are as follows: a. In the fiscal year 1980 (October 1979 to September 1980) there were 31 shelter programs serving “6 counties in the state. This number was unchanged for fiscal year (FY) 1981. b. There were 52,603 shelter days for women and their children in FY 1980 and 66,991! in FY 1981. c. There were 1,336 women denied shelter in FY 1980, and a 27.11 percent increase in FY 1981. d. There were 17,778 crisis calls in FY 1980 and 20,232 for FY 1981. These statistics are based on hand tallies and it is anticipated that the actual figures are larger. ln assault cases, the victim is predominantly the wife, but, in homicide cases, husbands are victims almost as often as wives. In a staff report to the National Commission on the Causes and Prevention of Violence, this fact was partially explained on the basis of aggression versus self-defense. While assault by men is an act of aggression, women commit murder in self-defense. In light of these statistics, it is amazing and unfortunate that there has been no conclusive research describing the battered woman herself. Yet without this information, it remains difficult for clinicians to identify this high-risk population, and lack of clarity impedes adequate preventive counseling. In a review of pertinent literature, only a small number of articles were found that hypothesize about the personality characteristics of the battered woman. While employing little statistical evidence, these articles suggest that she is isolated, withdrawn, dependent, and feels deserving of punishment. Both the small amount of data and the hypothetical nature of the available data support the need for this study. The study will attempt to pinpoint some of the personality characteristics of the battered woman. It will make objective data available to professionals in the field and will attempt to distinguish the battered woman from the nonbattered woman with respect to characteristics like differential acceptance of the feminine sex role, life history differences, and personal and marital adjustment. It is hoped that results of this study will assist in identifying those women who are likely to be at high risk if they place themselves in a potential battering situation. Purpose The purpose of this study is to explore several personality and environmental characteristics theoretically linked to battered women. The investigation will cover three primary areas: differen- tial acceptance of the feminine sex role, life history differences, personal and marital adjustment. Research Questions and Hypotheses This study will be based on the following research questions and hypotheses: Research Question 1: Are battered women oriented to more traditional sex roles than nonbattered women? Hypothesis: Battered women are more likely than nonbattered women to have traditional sex-role preferences. Research Question 2: Do battered women differ in their life histories from nonbattered women? Hypothesis a: Battered women are more likely than nonbattered women to have felt closer to their fathers than to their mothers. Hypothesis b: Battered women are more likely than nonbattered women to have experienced violence in their childhood home. Hypothesis c: Battered women are more likely than nonbattered women to have mothers who had traditional female sex role preferences. ijothesis d: Battered women are more likely than nonbattered women to have had fathers who had traditional male sex role preferences Hypothesis e: Battered women are more likely than nonbattered women to view their parents as exhibiting greater sex-role enforcement. Hypothesis f: Battered women are less likely than nonbattered women to have experienced family protectiveness. Hypothesis 9: Battered women are less likely than nonbattered women to have experienced family harmony and mother supportiveness. Hypothesis h: Battered women are less likely than nonbattered women to have experienced mother and father democracy. Research Question 3: Do battered women differ from nonbattered women in their personal adjustment? Hypothesis 3: Battered women are more likely than nonbattered women to report sexual difficulties and dissatisfaction. Hypothesis b: Battered women are more likely than nonbattered women to feel dependent. Hypothesis c: Battered women are more likely than nonbattered women to feel incompetent outside of the home. ijothesis d: Battered women are more likely than nonbattered women to perceive themselves as deserving punishment. Hypothesis e: Battered women are more likely than non battered women to be socially and emotionally isolated Hypothesis f: Battered women are more likely than nonbattered women to be unemployed outside of the home. Hypothesis 9: Battered women are more likely than nonbattered women to report having fewer outside interests. Hypothesis h: Battered women are more likely than women to have areas of conflict with Hypothesis i: Battered women are more likely than women to be submissive. Hypothesis j: Battered women are more likely than women to have lower self-esteem. Hypothesis k: Battered women are more likely than women to feel anxious. nonbattered their husbands . nonbattered non batte red non battered CHAPTER II REVIEW OF LITERATURE When this research began, literature concerning domestic abuse, and specifically battered women, was sketchy at best. A review of Psychological Abstracts for the twelve years between 1969 and 1981 yielded few references. Two separate bibliographies1 pro- vided most of the resources that specifically related androgyny to battered women. There are two distinct problems encountered by researchers on battered women. First, there is great difficulty obtaining a repre- sentative sample. Too few women are willing to admit to the problem and/or be subjected to study. Schuyler (1976) stated in her article on wife abuse that "women have so effectively hidden the abuse to which they have been subjected that there is an absence of direct and conclusive documentation of the scope of wife abuse" (p. 488). In a study by Flynn (1977), a sample of 14 women was drawn from an agency similar to a community mental health center. Flynn stated that "the strategy of obtaining a random sample of victims of assault was rejected . . . because it was assumed that few people would 1Annotated Bibliography of Woman Battering, compiled by Claudette McShane, 1977; Spouse Abuse: An Annotated Bibliography, compiled by Catherine Abramson, 1977. 7 readily admit that assaults were part of their family interaction, even if the participants were identified" (p. 15). Second, samples are biased by socioeconomics and by class with most subjects coming from lower socioeconomic groups. Women from upper strata hide abuse. Martin (1977) suggested that "the popular assumption by the middle class that marital violence occurs more frequently in the ghetto and among lower class families reflects the inability of middle-class investigators to face the universality of the problem" (p. 20). In reviewing the literature, most researchers use actual physical abuse as the defining characteristic of the battered woman. Gayford (1975) started the trend by describing the battered wife as a "woman who has received deliberate, severe and repeated beatings at the hands of her husband or lover and has suffered severe physical injury as a result" (p. 124). Many descriptions and explanations of the reasons for battering or being battered have been advanced in the literature. While these various theories each have their own unique aspect, they are also closely interrelated, and the outline below attempts to characterize, without unrealistically and rigorously separating, these concepts . Sex Role Socialization Sex role socialization in western society and, more specifi- cally, in the United States, may play a major part in the psychologi- cal makeup of the battered woman. Many of the problems faced by the battered woman may be related to inadequate childhood prepara- tion in self-reliance. Barry, Bacon and Child (1967) did a cross- cultural survey of sex differences in socialization and found that women are subjected to more pressure to be nurturant, obedient, and responsible whereas men are pressured to be achieving and self— reliant (p. 328). Women are allocated tasks at or near the home that emphasize their major role of ministering to the needs of others, i.e., cooking, cleaning, water carrying (Barry, et al., p. 329; Bem, 1974, p. 156). Although childbearing is biologically the role of women, child care has been socially assigned to them. The responsibility of childcare requires more nurturant behavior and more continuous responsibility than the tasks carried out by men, thereby keeping them other directed. Men, on the other hand, traditionally participate in the economic sphere or in warfare, which call for greater self-reliance and a higher degree of survival skills than those carried out by women (Barry, et al., pp. 329-331). The magnitude of the effects of sex role stereotypes that exist in American society was dramatized by a study of mental health clinicians by Broverman, et al. (1970). The study examined attitudes of male and female clinicians regarding the attributes characterizing healthy adult men and women, on the one hand, and healthy adults, sex unspecified, on the other. It was found that clinicians have different concepts of health for men and women, and that these differences parallel the sex role stereotypes that exist in American society (p. 5). The concepts of a healthy man were found not to differ from those of a healthy adult whose sex was unspecified, 10 but the concepts of a healthy woman held by the clinicians did differ significantly from those of healthy sex-unspecified adults. Based on these findings, Broverman et al. stated: . for a woman to be healthy from an adjustment view- point, she must adjust to and accept the behavioral norms for her sex, even though those behaviors are generally less socially desirable and considered to be less healthy for the generalized competent mature adult (p. 6). For centuries the primary domain of women's achievement has been preserving family unity. Many times a woman carries with her an idealized adolescent conception of the feminine role that comes into direct conflict with her adult appraisal of reality (Ball, 1977, p. 4). This conception of the feminine role carries with it the belief that love will give a feeling of worth and meaning to her life. It is for this reason that many women, who also believe that marriage and children are expected of them, are unable to perceive alternative lifestyles and rush into disastrous marriages (Martin, 1977, p. 76). Since the woman's sense of achievement comes from the family unity it becomes crucial that violence be denied and that family life be portrayed as happy (Steinmetz 8 Strauss, 1973, p. 52; Walker, 1978, p. 530; Rounsaville, 1978, p. 16; Hilberman, 1980, p. 1342). Women who accept the traditional responsibility for the success of their marriages feel guilty when their marriage turns violent_(Prescott 8 Letko, 1977, p. 80). The ”womanwalso‘feuarsflthat , ifthe battering—“situationbecomes known outside of the home it will be perceived as a failure on her part to maintain a healthy family atmosphere (Schuyler, 1976, p. 490; Walker, 1978, p. 530). 11 The battered woman, if highly sex-typed, has become motivated during the course of sex role socialization to keep her behavior consistent with an internalized sex role standard (Bem, 1975, p. 634). Additionally, the highly sex-typed woman is only able to keep her behavior consistent with her internalized sex role standard by suppressing any behavior that may be considered undesirable or inappropriate for her sex (Bem, 1974, p. 155). Therefore, the weaker feminine sex role behaviors are maintained, while stronger behaviors associated with the masculine sex role are suppressed. The fact that women are socialized to give nurturance __‘ n offers some explanation for the _'_'oh-but-he-needs-me" syndrome “"~-. \ found in the battered woman who feel‘svacom'plete or powerful only when ”caring for another person. I The effects of sex role socialization on men may also contri- bute to the battering situation. Flynn (1977) indicated that the batterer has certain sex-role expectations of his wife that may not be met, i.e., that her personality or behavior should not change, that she should be submissive, and that she should not cope well with difficulties. When the woman proves better at verbal arguments the man switches to physical skills in order to dominate, which punishes or defeats any attempts at strength on the woman's part (p. 18). There is also the tacit assumption operating in violent families that a man's wife is his property to do with as he pleases. This has been perpetuated through social customs that deprive women of full, independent adult status (Schuyler, 1976, p. 489). 12 The idea that a narrowly defined feminine self-concept may inhibit behaviors that are stereotyped as masculine may account, in part, for the passivity and submission patterns observed in battered women. Lacking one half of an effective behavior repertoire the battered woman has a low self concept and, at times, may despise herself. By only having one set of behaviors to draw from the battered woman finds herself in a subordinate position with respect to her husband. Several studies have stressed that violence most frequently occurs between people who can be differentiated on superior-subordinate poles (O'Brien, 1971, p. 693; Sharness, 1977, p. 115; Schuyler, 1976, p. 489). It has also been suggested that family violence most commonly involves the use of physical force on the part of the husband, who is either underachieving in the worker-earner role or is a potential underachiever in comparison with his wife (O'Brien, 1971, p. 695). Violence becomes his method of maintaining superior-subordinate roles, and, in turn, the dependency of his wife (Hilberman, 1980, p. 1338). Since the battered woman with fragile self-esteem depends on the esteem of her husband, submission to physical violence may be a trade-off for maintaining his approval (Schuyler, 1976, p. 489). Learned Helplessness Several authors in recent years have pointed to the concept of learned helplessness as an explanation for the passivity and inability to effect change felt by many battered women. When a woman/girl has experienced consistent powerlessness or trauma she 13 cannot control, her motivation to respond to later trauma wanes. Moreover, even if she does respond and the response succeeds in producing relief, she has trouble learning, perceiving and believing that the response worked. Finally, her emotional balance is dis- turbed. Depression and anxiety, measured in various ways, pre- dominate (Seligman, 1975). Initially the woman believes that a harmonious relationship is not within her control and submits to beatings in order to avoid further abuse. Any attempt to bring about change is met with further beatings, resulting in a passive and fatalistic approach to current and future problems (Walker, 1978, p. 525; Walker, 1980, pp. 45—51; Ball 8 Wyman, 1978, p. 546; Hilberman, 1980, p. 1343; Follingstad, 1980, p. 294). Morbid Dependency Conceptually, "morbid dependency"_ as described by Horney (1967) appears to play a large part in the dynamics of battered women. Morbid dependency is based on the idea that to love means to lose oneself, to submerge oneself in ecstatic feelings and to merge with another in order to find unity (Horney, 1967, p. 240). It implies a removal of feelings of guilt, elimination of responsibility for self and of the individual's struggle with a harsh world for which she is not prepared. The morbidly dependent woman seeks love in the hope that there will be no more loneliness, no more feeling lost. Love gives her a feeling of worth and meaning to life, as well as a sense of salvation and redemption. She is spellbound by persons of the opposite sex who impress her as stronger or superior (p. 293). 14 She overrates the other person who appears to possess attributes she bitterly misses in herself (p. 241). Physical abuse inflicted by her mate, therefore, is often interpreted as failure and she sees any friction that arises as being her fault. She believes that she should be more serene, more thoughtful, more forgiving (p. 241). Submission to physical violence is often seen as a trade-off for maintaining her husband's approval (Schuyler, 1976, p. 489). Two imperatives which result are: (1) I should be able to develop any love relationship into a state of absolute harmony, and (2) I should be able to make my partner love me (Horney, 1967, pp. 240-242). For some women, this means becoming isolated from others and depending solely on their husbands for a sense of self and self-worth . Isolation Along with morbid dependency, one must consider social isolation as a major component of the battering situation. Hanks and Rosenbaum (1977, p. 305) and Hilberman (1980, p. 1339) found that almost all couples studied were socially isolated and had few, if any, other close relationships. Neither partner could tolerate the other having close relationships outside the marriage. The women feared that male friends of the husband would encourage his infidelity; the men feared that female friends of the wife would encourage her leaving. This social isolation begins a vicious cycle in which the woman feels depressed, becomes increasingly dependent on her 15 husband, and thus finds it difficult to leave. Depression and despair are increased as the woman attempts to improve the situation and fails. She feels trapped and sees attempts to free herself as futile (Martin, 1977, p. 84). She fears loneliness and economic hardship, believes that divorce is stigmatized, and doubts that she can get by alone (Gelles, 1976, p. 660; Hanks 8 Rosenbaum, 1977, p. 302). In addition, it has been found that social isolation is imposed on the battered woman by her husband who is jealous and controlling. This social isolation is reinforced by (1) the woman's fear of further suspicions and compliance with her husband's restric- tions, (2) her fear that if she leaves home her injuries will be obvious to the public, and (3) her expectation that no one will believe her story, and that she will be blamed for whatever has occurred (Goodstein 8 Page, 1981, p. 1040). Violence in Families of Origin There is much debate but little research regarding family history of battered women. Some studies have indicated that battered women suffered family abuse and neglect in childhood (Ball, 1977, p. 4; Gelles, 1976, p. 663; Hilberman, 1980, p. 1338). Hanks and Rosenbaum (1977) found that women who stayed with violent, alcoholic-abusing men had mothers who had been protective toward men and were uncritically overprotective of their sons. On the other hand, these women had fathers who served a dual role of punisher and advocate. The father administered the punishment and bargained 16 with the wife/mother for lesser punishment (Hanks 8 Rosenbaum, 1977, p. 293—94; Goodstein 8 Page, 1981, p. 1038). Gelles (1976, pp. 662-63) has hypothesized that violence in the family of origin may present a model for the victimized woman. Some women may have grown up with the expectation that husbands are "supposed" to hit wives. He reported that women who grow up in surroundings which include and approve of family violence are more likely to marry men who are prone to use violence. In his study of 80 families, Gelles (1977) stated: Women who had observed conjugal violence in their families of orientation were likely to be victims of marital violence in their families of procreation. Not only does experience with and exposure to violence as a child influence whether a woman is likely to be hit by her husband, experience and exposure to violence affect what a woman does after she has been hit (p. 60). Note the relationship between this and learned helplessness discussed earner. Prescott and Letko (1977, p. 83) found "fighting" between parents in 24 out of 30 childhood homes of battered women. Hanks and Rosenbaum (1977) found that women who both witnessed and received physical abuse as children come to accept physical abuse as inherent in male/female relationships and were not disturbed by its occurrence (p. 302). Further support is offered by Goodstein and Page: Parallels could be seen between the current lives of these women and their childhood parental interactions. Their mothers were controlling of bombastic but ineffectual fathers, and now the daughters married men who often needed rescuing. Their fathers severely punished them; now their husbands served that role (1981, p. 1038). 17 It is interesting, as well as representative of the lack of conclusive findings in this area, that Roy (1977, p. 3) indicated that women who have experienced abuse as children and/or parental fighting expect it in the marriage and are less likely to seek help. In her study of 150 cases of battered women who responded to an Abused Women's Aid in Crisis Hotline questionnaire it was found out that few among them came from violent childhood homes (p. 31). Perhaps because their childhoods were not violent they were less likely to tolerate violent situations. Contradictory evidence was presented by Gelles (1977) who found that women who had experienced childhood violence are more likely to seek help and are less tolerant of family violence (p. 60). The Three Phase Cycle of Violence The most recent literature has indicated that the violence found in wife battering is actually comprised of three distinct phases: (1) the tension building phase, (2) the acute battering incident, and (3) the kindness and contrite loving behavior phase. Phase 1, the tension building phase, is marked by minor battering incidents and rapidly increasing tension. During this phase the woman attempts to calm the batterer by employing tech— niques that have been helpful in the past, such as becoming extremely complaint, more nurturing, and attempting to provide the batterer with whatever he desires (Walker, 1980, p. 56). The woman also lets the batterer know that she is deserving of the abuse. This is done primarily to prevent his anger from escalating. In order to 18 remain in the passive, nurturing and compliant role the woman denies her own anger and covers it over with passivity (Hilberman, 1980, p. 1342; Walker, 1981, p. 85). Additionally, the woman attempts to control all external events that would cause the batterer to explode. This is an extremely. difficultTF'rf'notan impossible, task. If each battering incident can be blamed on an external event she can blame that event and not the batterer, thereby making it easier to deny her own anger (Walker, 1980, p. 57). The result is that the woman's tension increases with that of her husband. As a result of the learned helplessness syndrome, discussed earlier, she feels that she is powerless to prevent the rest of the cycle from occurring. Phase 2, the acute battering incident, is characterized by the discharge of Phase 1 tension. During this phase the batterer's rage is so great that it blinds his control over his behavior (Walker, 1980, p. 60). It is this lack of control and its major destructiveness that makes this beating different from the minor incidents that existed in Phase 1 (Walker, 1980, p. 59). It is impossible to predict the type of violence that will occur. During the onset of Phase 2, the anticipation of what might occur causes the woman to become anxious and depressed as well as to experience other psychophysiological symptoms (Walker, 1980, p. 61). While the battering incident is occurring the woman knows what to do to prevent inciting the batterer further; she remains calm and waits for the storm to end. In order to do this the woman may dissociate: 19 Some women say that it was as though they could stand back and watch their disembodied selves being thrown against a wall or down a flight of stairs. The dissociation is coupled with a sense of disbelief that the indicent is really happening to them (Walker, 1980, p. 62). Phase 3 is the phase that both the battered woman and the batterer look forward to. The batterer behaves in a loving manner, begs her forgiveness and promises that it will never happen again (Hilberman, 1980, p. 1342; Walker, 1980, p. 65). Since the battered woman is invested in keeping the family together as well as believing that one of her major roles in life is to be a caretaker she is willing to believe the batterer's promises. This behavior on the batterer's part serves as the woman's reinforcement for staying in the relation- ship (Walker, 1980, p. 67). mm The battered woman is still not clearly understood. Theoretically what appear to be common elements are that she is dependent for both financial and emotional reasons and seeks her husband's approval. She has attempted to fulfill the more traditional aspects of the female role by becoming devoted to family and/or husband. She is nurturer and caretaker to an extreme. She stays in the subordinate position where violence is likely to be inflicted by her husband who is in a superior position. The battered woman is also isolated from others and believes that to be in a violent marriage is not unusual. This isolation leads to increased dependency. Parental modeling is also a basis for tolerating abuse. CHAPTER III THE QUESTIONNAIRE History of the Questionnaire A pilot questionnaire was developed approximately two years before the final data collection. Subsequently, the questionnaire was revised four times. The first pilot questionnaire consisted of a series of open- ended questions. It was distributed to approximately 30 women, some battered and some not. Analysis of the questionnaire indicated that the open-ended responses were too difficult to code. A search was begun to locate research instruments in other studies of domestic abuse. Several questionnaires were found; how- ever, they dealt primarily with non-psychological issues like marital relationship, extent of injury, and whether or not the victim obtained help at a hospital. The original open-ended questionnaire was then reconsidered because of its psychological intent. It was simplified for easier coding and the second pilot study was begun. The purpose was to re— evaluate the wording of the questionnaire, which originally was found to be too sophisticated, repetitive and biased toward negative responses. It was again found to be too difficult to code. Conse- quently, it was decided to use a forced-choice format. 20 21 To assess role acceptance, Janet Taylor Spence and Robert Helmreich were contacted about their current research involving androgyny. Two instruments, the Parental Attitudes and Parental Attributes scales were incorporated into the questionnaire for this study. Items for the final questionnaire were phrased in forced choice fashion, and then refined by removal of remaining duplication. Seven individuals were asked to categorize the items under the head- ings of dependent variables. When four or more persons rated an item under the same dependent variable it became an item for that variable. If there were fewer than four agreeing, the item was dis- carded. A final pilot study was run in order to smooth out the remaining rough spots in the instrument. The questionnaire was given to 15 women of a socioeconomic and educational level similar to that of the women living in the shelter home. Questions which were found to be confusing were eliminated. In addition, questions which contained the words "don't" and "no" were reworded in order to read more positively. Finally, two scales were added to the questionnaire. The subscale for personal respect from the Rotter Locus of Control Inventory was added in order to complete the scale for "Deserving of Punishment." The Bem Sex Role Inventory was added to further indicate Sex Role Acceptance. The final questionnaire appears in Appendix A. 22 General Description of Content The questionnaire was designed to assess three major areas in the lives of battered women: (1) Sex Role Preference, (2) Life History, and (3) Personal Adjustment. The outline which follows briefly describes the scales and items included in each of the sections of the questionnaire. l. Sex Role Preference: A. Assessment of feminine sex role identification. The measure used is the Bem Sex Role Inventory. This measure is com- prised of three subscales named Masculinity, Femininity, and Neutral with a method of computing androgyny. The items in the inventory are those items which have been found to be characteristic of masculine and feminine sex roles. ll. Life Histogy: A. Personal history information which includes a history of physical and sexual abuse as well as demographic data. The demographic material includes age, number of children living with the woman, years of school completed, type of job held, marital status, husband's age, education and occupation. There are also several items to determine religious affiliation and importance. Many of these items are common to question— naires which collect basic demographic data for use in describing the sample. Items which are less common, such as those directly related to family violence, were adopted from the previously described pilot project and from the Abused ‘l ,4 ‘11' :‘ 'Wlf‘_ .. . ’ ”fay,- l V 1' 'on 'r. ‘93.... 23 Women's Aid in Crisis, Inc. Short Form Intake Questionnaire (Roy, pp. 307- 308) . Assessment of parental attitudes toward the subject. Included in this are scales measuring closest parent, father positivity, mother positivity, sex-role enforcement, family protectiveness, female family harmony and mother supportive- ness, mother democracy, and father democracy. Assessment of parental attributes along the masculinity- feminity dimension. The scales included here are mother femininity, mother msaculinity, father femininity, and father masculinity. Personal Adjustment: A. Assessment of self-concept. Included in the assessment of self-concept are scales for dependency, competency/ incompetency outside the home, personal respect, guilt, deservingness of punishment, social/emotional isolation, employed/unemployed status outside the home, and outside interests. Assessment of the marital relationship. Included in the assessment of the marital relationship was a devised scale for conflict, as well as scales from the Dyadic Adjustment Scale which evaluated marital concensus, affectional expression, satisfaction, and cohesion. 24 Item Development The development of items devised by the author of this study has been discussed under the topic "History of the Question- naire." At this time a brief history of the Bern Sex Role Inventory, The Parental Attitudes and Attributes Scales, the Dyadic Adjustment Scale, and the Rotter Internal-External Locus of Control Scale will be given. Refer to Table 3.1 (pp. 30-32) for a compact analysis of the scales, and corresponding hypothesis numbers, measures, items and established and determined reliabilities. Established reliabilities are the predetermined reliabilities, determined reliabilities are those found for this study and its specific sample. I. Sex Role Acceptance: The items measuring sex role acceptance are taken from one source, the Bem Sex Role Inventory, and provide a measure of masculinity, femininity and adrogyny. Bern Sex Role Inventory (BSRI).-—A major scale used for this study was the rating on the femininity scale of the Bem Sex Role Inventory (BSRI). This scale was used in order to provide insight into whether battered women align themselves more with the traditional feminine sex role than non-battered women do. The inven- tory also provides a masculinity score and an adrogyny score which will be used to assess purityof role preference. The BSRI is a unique scale, having been based on the theory that a sex-typed person has internalized society's sex-typed standards of desirable behavior for women and men but has not 25 based these standards on differential endorsement by males and females. Both the masculinity scale and femininity scales are com- prised of 20 personal characteristics each. The individual is asked to indicate on a seven-point scale how well each of the items (feminine, masculine, and neutral personality characteristics) describe herself. Each subject receives three major scores: 1. a femininity score--a self-rating for all endorsed feminine items. Reliability = .90 Number of Items = 20 2. a masculinity score--a self-rating for all endorsed masculine items. Reliability = .90 Number of Items = 20 3. an androgyny score--which "reflects the relative amounts of masculinity and femininity that the person included in his or her self-description . . . it best characterizes the nature of the person's total sex-role" (Bem, 1974, p. 258). The androgynynous score is the subject's ratio for the difference between her masculine and feminine self-endorsement. This is done to ascertain if her endorsement of masculine attributes differs signifi- cantly from her endorsement of feminine attributes. This allows classification of the woman as sex-typed or androgynous (Bem, 1974, p. 158). This score is useful also in that it allows a comparison of different popula— tions with respect to the percentage of significantly sex-typed individuals existing within each (Bern, 1974, p. 158). High positive scores indicate masculinity. The closer the score comes to zero, the more androgynous the individual. ll. Life Histories As previously stated, many of the questions of a demographic orientation are similar to those used in other questionnaires. Questions pertaining specifically to abuse were adapted from the Short Form Intake Questionnaire used by Abused Women's Aid in Crisis, Inc. 26 Those items which dealt specifically with parental attitudes and attributes were taken from the scales devised by Janet T. Spence and Robert L. Helmreich and can be found in the book Masculinity and Femininity: Their Psychological Dimensions, Correlates and Antecedents. Parental Attitudes Questionnaire.—-This questionnaire consists of two sections. The first contains 58 items that examine parents' attitudes and behaviors. The first eight items pertain to the family as a whole. These are followed by parallel statements, 25 each, about the mother and father. Each item is measured on a five-point scale, ranging from "very characteristic" to "very uncharacteristic" (Spence 8 Helmreich, p. 40). The last part of the questionnaire was developed to determine the parent to whom the respondent feels closest or who she most resembles in ideals and personality. These items are scored separately based on the five response alternatives which follow each euestion. The intent of these questions was to provide a measure of mother versus father identification. In devising this questionnaire Spence and Helmreich con- sulted interview schedules and objective questionnaires used pre- viously in other investigations of parent-child relations. In identify- ing items used by other investogators (Baunrind, 1971; Coopersmith, 1967; and Heilburn, 1973) to evaluate family closeness and harmony, the behaviors found to be most critical were degree of affection, use of praise, criticism, and punishment as disciplinary techniques, use I .. 7815.. . f ' 1 I 0. o (0103'. I . ’rfiibt ill.‘ Pu '...-. AQJL‘ 27 of reasoning, strictness of rules, consistency of rule enforcement, and degree of encouragement of independence. These were found to have a major impact on the development of children (Spence 8 Helmreich, p. 41) . Parental Attributes Questionnaire.—-Spence and Helmreich had initially designed a Personal Attributes Questionnaire (PAQ) and then asked their subjects to rate their parents on those same items. A brief description of the Personal Attributes Questionnaire is given here so that the reader has an understanding of the items used in the Parental Attributes Questionnaire. The Personal Attributes Questionnaire is a self-report instrument with each item placed on a bipolar scale. Items used are believed to be characteristics that differentiate the sexes (Spence 8 Helmreich, p. 19). It may be noted that these items are 55 bipolar items "drawn from a pool of over 130 items that largely had been put together by Rosenkrantz, Vogel, Bee, Broverman and Broverman (1968) from nominations by students of characteristics differentiating men and women" (Spence 8 Helmreich, p. 32). The chosen items were selected from those for which both sexes demonstrated a con— sistent stereotype about sex differences. The PAQ "is made up of items describing characteristics that are not only commonly believed to differentiate the sexes but on which men and women tend to report themselves as differing" (Spence 8 Helmreich, p. 32). The Parental Attributes Questionnaire is composed of 16 items selected from the above PAQ and broken into scales of mother 28 femininity, mother masculinity, father femininity and father mascu- linity. Items were chosen on the basis of their magnitude of the part-whole correlation between the item and the scale to which it belonged (Spence 8 Helmreich, p. 35). The subjects were asked to rate their mothers as they perceived her, or if they had been raised most of their lives by a step-mother, foster mother, or other female guardian, they were asked to rate her accordingly. The same pro- cedures were used with fathers. lll. Personal Adjustment Questions used to answer hypotheses regarding personal adjustment were taken from two sources. Those evaluating such variables as dependency, competency/incompetency, social/emotional isolation, employed lunemployed outside of the home, outside interests, submission/dominance, self-esteem and anxiety were devised by the author during the pilot study. Items used to respond to variables regarding the marital relationship were a combination of author devised items and previously established scales. These variables included sexual satisfaction which included devised items as well as the Affectional Expression scale from the Dyadic Adjustment Scale. The variable deserving of punishment was a combination of author devised items and items dealing with Personal Respect taken from the Rotter Locus of Control. Items responding to the level of con— flict within the marital relationship were a combination of author devised conflict items and the intact Dyadic Adjustment Scale. .16 no .r.‘ :.o‘;> ~’ jl“.hol I u .L‘ - I ‘ ”I. r fl . 5.;L".‘ ‘ r." :” l ' , -°‘..."n,..-u 29 Dyadic Adjustment Scale.--The Dyadic Adjustment Scale developed by Graham B. Spanier was used in this study for assessing the quality of marriage and similar interpersonal dyads, i.e., married or unmarried cohabiting couples. This scale is comprised of 32 items which make up four subscales: (1) dyadic satisfaction, (2) dyadic cohesion, (3) dyadic concensus, and (4) affectional expression. In the purpose of scale development Spanier and Cole viewed dyadic adjustment as a process rather than a static state (Spanier, p. 17). Since dyadic adjustment is viewed as a process, the scale evaluates the relationship at a point on a dimension from well- adjusted to maladjusted. Internal versus External Control of Reinforcement Scale.-- This scale was devised by Julian Rotter in 1966. It measures the individual's belief about who controls her behavior and life events. An individual who has an internal locus of control believes that she is self—controlled, while the individual with an external locus of control perceives control as coming from outside of herself. It has been found that internals take more steps than externals to improve environmental situations. Externals need a structured environment. A subscale of I-E, personal respect, was used in this study (Rotter, 1966) . 30 I I 8. -- 9. .2. .2 .2 .2 .2 .2 k .z 3S: 528.50 .5522 9. .9. .3 .2. .2 .2 .3. .93: 3653:8833 I I am. I .3 .2” .2 .2 .mN .2 .3 .m. .o. 3. mm 8522 fine >553: >=Emu Beep“. I I R. I 8 .3. .2 m .N 3S: 8853689.“. 2.5.... I I S. I N... .3 .3 .2 e em 3,}: 28588.5 8.3. x3 I I a... I S. .2: .8. .8. .8 .5 .3 .3 a B 35$: 3.5558“. .653. I l 3. I 2: .3. .2: .8. .8. .8 .mm .2 a B 35$: 3.5.38.2 85m“. I I S. I N... .2 .3. .B .3 .3 .2 .K a UN 35$: 32:38: 8522 I I 3... I a .8 .3 .3. .S .3 .8 .2 a 8 33.2.: 3:555". LBS: I I E. I S. .3. .... .8. a £ .8259 3533 89.2.5 I l a... l 2. .9. .3 .5 .R .3 .3 .on a 3 3S: 3328.. Lafiez I I 8. l S a... .8 .5 .8 .mm .3 .3 .3 .S .. 2 3,}: 2.3.7.8 L653. I l 2. I on .m. .i .N. a - 3%: 3:83. .886 a.~ .e.~ .m.~ .e.~ ..e~ .eeN ..8~ .am~ .mmN .NmN .eeN .e=~ I I 3. 8. .mi .2; .52 .32 :2 .82 .32 .-~ 2 a. :32 3.5.38: omN ...~ .=.~ ...~ weN .meN .NeN .mmN .omN .mm~ .emm ..a~ I l 3. 8. .3; :2 52 5.2 .~2 62 .32 .RN 8 m. :32 3.2.58“. 8.8.80 8.2.5 $33328”. :>.__Be_.em 88:52 .5: use: .8532 283 89.52 932 “805.5630 pocm__om3w Cc 330509»: Ed: and? Lon—:32 .mu_._:om__mm Dam .mconEaz Eu: .mLuDEJZ momunuoa>z .muEmz o_mumi...m w4mo_QEo:D\Oo>o.QEm c- 5.... I I S. I .3. .2.. .o... .3. .mm. .mm. .2. a on 2.3.50. c0223.. .55”. 250300.... :2. a: . Um 2.3.3.0. .chcmEam ..o 0...)..030 m. 3. .... I .3. .mm. .m. m R 33300. :30 2.52.22. 3... .2. I m... .2 .3. .2. .2. .2. a R 22.50 B .38.. .833. .38.... a... I S. .3. .2. m on 333.5. 38.363.35.358... a 8. a... I 3. .3. .8. .a .. m. a. 63.59 58.2.33 I I m... I 3.2 .22 .3. .2: .. an 3.5. 20.3955 .2288... I I a... I a... .5. .8. .3. .. mm .2333. 8:83.33 23.83 I I on. I on .3 .3 .3 .3 .3 .9... .3. o .2 .m.._5m:am ..>.._5m:om 3.5.5.2 EB. 3.5.. 39.52 «.33 #5532 9.33 855.3..0 393.333... .0 339.69»... 9.02 30:52 .5: -Irll .Umac_.coUI .. .m m..m<... 32 .ommucou..oo. new “:38 3:88. a .659. :5 32:52.9. 855.03... 9.. m. v.55 p.253 .5» 3.3m Em: ace 3 9.80.. .550 >ucoaaaun. .>. 339.0 c. flown-Um? m. m7: fie.“ .0363 ”5.8% m2... 9.6 >333 m7: .3. 0:38 3.3.32.3. $032.93.”; “558330 iCfi .23... .9...th .o 52.0.33. >9 355.33.. 3.2.52.3. nm.m.:...sm..m.§. Uocm:nfimw .1 28m 5953.3 2.8.5 u 93 # 0.3m m2:£..3< 333?. u m..=z ES. 932 32:52 HEM. . l!l| .Baczcou--...m Mimi 33 Definition of Variables This section will present both a definition of variables and the definition of the scale or scales used to measure the variable. Independent Variable The independent variable has two components. (a) Battered Woman -- a subject in this study. For the purposes of this study, she was defined as a woman who had been physically mishandled on more than one occasion by another person with intent to cause bodily injury, the other person being a man with whom the victim was living at the time of the battery and with whom she was engaged in a primary relationship. (b) Non-battered Woman -- a subject in the control group in this survey study. For the purpose of this study she was defined as a woman who did not indicate having been physically hit or otherwise physically mishandled by the man with whom she was living. In order to assess whether or not a woman would be con- sidered battered, Question 116 was placed in the questionnaire: 116. As an adult, I have been physically mishandled or abused by my husband: Very Often Often Occasionally Seldom Never 34 Dependent Variables SECTION I: SEX ROLE ACCEPTANCE A. Variable: Measure: Scales: SEX ROLE PREFERENCE -- an individual's sex role acceptance was her endorsement of masculine or feminine attributes. Bem Sex Role Inventory (BSRI) (a) Femininity score--a self-rating for all endorsed feminine items. Established Reliability Determined Reliability Number of Items = 20 Item Numbers: 223, 226, 229, 232, 235, 238, 241, 244, 247, 250, 253, 256, 259, 262, 265, 268, 271, 274, 277, 280 .90 .85 (b) Masculinity score--a self-rating for all endorsed masculine items. Established Reliability = Determined Reliability = Number of Items = 20 Item Numbers: 222, 225, 228, 231, 234, 237, 240, 243, 246, 249, 252, 255, 258, 261, 264, 267, 270, 273, 276, 279 .90 .91 SECTION II: LIFE HISTORIES A. Variable: Measure: Scales: CLOSEST PARENT -- this variable was indicative of the parent with whom the individual aligned herself. Parental Attitudes Scale (a) Closest Parent--this scale provided a measure of mother versus father identification and was designed to determine the parent who the woman felt closest to or who she most resembled in personality and ideals. Determined Reliability = .73 Number of ltems = 4 Item Numbers: 72, 74, 75, 76 B. Variable: Measure: Scale: C. Variable: Measure: Scales: 35 (b) Father PositivitynFather was encouraging and supportive, consistent, physically affectionate, used praise, and used reason (Spence 8 Helmreich, p. 42). Determined Reliability = .83 Number of Items = 11 Item Numbers: 49, 52, 53, 54, 55, 60, 61, 63, 66, 69 (c) Mother Positivity--Mother was nonpunitive and noncritical, encouraged discussion, was interested and consistent; there were few arguments (Spence 8 Helmreich, p. 42). Determined Reliability = .88 Number of Items = 8 Item Numbers: 30, 33, 34, 37, 41, 43, 45, 46 CHILDHOOD VIOLENCE -- this variable was defined as the woman having experienced physical and/ or sexual abuse during childhoos, or having observed the mother being physically abused by the father. Devised ltems (a) Childhood Violence--Definition was the same as that for the variable. Determined Reliability = .57 Number of Items = 8 *Item Numbers: 109, 110, 111, 112, 113, 114, 115, 167 MATERNAL SEX ROLE PREFERENCE -- the mother's endorsement of masculine and [or feminine traits in herself. Parental Attributes Scale (a) Mother Femininity--the mother's endorsement of feminine traits in herself. Determined Reliability = .89 Number of Items = 8 Item Numbers: 78, 80, 81, 82, 84, 85, 90, 91 * Only Items 109, 111, 114, and 167 were used in the analysis. Items 110, 112, 115 are open-ended items used for further clarification. Item 113 was used for more specific information. 36 (b) Mother Masculinity-"the mother's endorsement of masculine traits in herself Determined Reliability = .69 Number of Items = 8 Item Numbers: 77, 79, 83, 86, 87, 88, 89, 92 D. Variable: FATHER SEX ROLE PREFERENCE —- the father's endorsement of masculine and /or feminine traits as perceived by the daughter (the woman in this study). Measure: Parental Attributes Scale Scales: (a) Father Masculinity--the father's endorsement of masculine traits in himself. Determined Reliability = .65 Number of Items = 8 Item Numbers: 93, 95, 99, 102, 103, 104, 105, 108 (b) Father Femininity-~The father's endorsement of feminine traits in himself. Determined Reliability = .88 Number of Items = 8 Item Numbers: 94, 96, 97, 98, 100, 101, 106, 107 E. Variable: SEX—ROLE ENFORCEMENT -- father and mother were unsympathetic to "women's lib," and minded child playing with opposite-sex toys (Spence 8 Helmreich, p. 42). Measure: Parental Attitudes Scale Scale: (a) Sex Role Enforcement--definition as above for variable. Determined Reliability = .53 Number of Items = 4 Item Numbers: 27, 38, 51, 62 F. Variable: Measure: Scale: C. Variable: Measure: Scale: H. Variables: Measure: 37 FAMILY PROTECTIVENESS -- Mother and father were cautious about their child for fear of child being hurt, and most often knew what this child was doing (Spence 8 Helmreich, p. 42). Parental Attitudes Scale (a) Family Protectiveness--definition as above for variable. Determined Reliability = .48 Number of Items = 3 Item Numbers: 17, 44, 68 FAMILY HARMONY AND MOTHER SUPPORTIVE- NESS -- family was close, the parents were pleased by their child having a career, the parents encouraged the child to stick up for her rights; the mother was perceived as having been affectionate, praising and encouraging. The mother used reason with the daughter. The respondent wanted to bring up her children in the same way (Spence 8 Helmreich, p. 42). Parental Attitudes Scale (a) Female family harmony and mother supportive- (1) (2) ness. Determined Reliability = .87 Number of Items = 16 Item Numbers: 16, 19, 21, 23, 25, 28, 29, 30, 31, 36, 37, 39, 40, 42, 45, 56 MOTHER DEMOCRACY -- Mother was encouraging, used reason, did not insist on her own way, both the mother and father encouraged the questioning of rules (Spence 8 Helmreich, p. 42) FATHER DEMOCRACY -- Father used reason, was strict, encouraged discussion, was non- critical and allowed questioning (Spence 8 Helmreich, p. 42). Parental Attitudes Scale Scales: SECTION III: A. Variable: Measure: Scales: (a) (b) 38 Mother Democracyndefinition is the same as for variable above. Determined Reliability = .88 Number of Items = 7 Item Numbers: 24, 25, 28, 36, 37, 40, 46 Father Democracy--definition is the same as for variable above. Determined Reliability = .70 Number of Items = 8 Item Numbers: 52, 56, 57, 58, 65, 57, 59, 70. PERSONAL ADJUS TMEN T SEXUAL SATISFACTION -— when sexual satis- faction existed in the relationship, there was a minimum of sexual problems, the woman was active while having sex, there were open demonstrations of affection. Sexual Satisfaction was measured by (a) (b) (1) devised items for use in this study; (2) affectional expression scale from the Dyadic Adjustment Scale. Devised Questions--this measures degree of sexual problem, satisfaction in present sex life with the husband, level of activity for the woman while having sex, and whether or not the woman has been raped or sexually molested as an adult. Determined Reliability = .48 Number of Items = 4 Item Numbers: 131, 163, 164, 168 Affectional Expression--this subscale dealt with sexual satisfaction and expression of physical affection within the relationship. Extablished Reliability Determined Reliability Number of Items = 4 Item Numbers: 187, 189, 213, 214a .73 .85 B. Variable: Measure: Scales: C. Variable: Measure: Scale: D. Variable: Measures: Scales: 39 DEPENDENCY —- this variable related to the woman's ability and desire to exist financially outside of her marriage. It also related to her feelings of competency and assertiveness, or incompetency and passivity. Devised Dependency ltems (a) Devised Dependency ltems--the scale measured the characteristics described under Variable: Depencency. Determined Reliability = .48 (Item 9 included) .66 (Item 9 deleted) Number of Items = 5 Item Numbers: 8, 9, 150, 152 INCOMPETENCY OUTSIDE OF THE HOME -- this related to the woman's ease in keeping a job and general feelings of competency. Competency / I ncompetency (a) lncompetency/Competency--Devised Items-- @115 scale was composed of items designed specifically for this study and examined the characteristics described under Variable: Incompetency Outside of the Home. Determined Reliability = .44 Number of Items = 3 Item Numbers: 127, 150, 152 DESERVING/NON-DESERVINC OF PUNISHMENT-- this variable was related to the soman's opinion as to whether or not she did any thing to elicit her husband's abuse, in addi- tion it explores the amount of control she felt she had in the way people interacted with her. (1) Items from Locus of Control; (2) Devised items (a) Personal Respect (Locus of Control)--this scale measured the level of control the woman felt she had in the way people interacted with her. E. F. Variable: Measure: Scales: Variable: Measure: Scale: 40 Determined Reliability .15 (Items 124, 134, 143 included) .44 (Items 124, 134, 143 excluded) (b) Cuilt--Devised for this study and explored feeling guilty as well as a genersl sense of morality. Determined Reliability .42 (Item 15 included) .57 (Item 15 excluded) Number of Items = 3 (with Item 15 included) (c) Deserving of Punishment-~this scale is com- posed of one item from which a response frequency will be calculated. The item assessed the degree of responsibility the woman felt for the abuse. Item Number: 118 ISOLATION -- this variable related to the amount of contact the woman had with other people as well as the amount of time spent at home. It involved both social and emotional isolation. Devised Items (a) Social/Emotional Isolation--this scale assessed ease of making friends, shyness, loneliness, amount of time spent at home, like/dislike of weekends and vacations. Determined Reliability = .69 Number of Items = 9 Item Numbers: 125, 126, 133, 138, 140, 147, 151, 219, 220 EMPLOYED/UNEMPLOYED OUTSIDE THE HOME-- this variable related to whether or not the woman was working at a job outside her home. Devised ltems Employed/Unemployed Outside the Home--this scale was composed of one open ended questions: Type of job you now hold (job title): C. Variable: Measure: Scale: H. Variable: Measures: Scales: 41 The responses to this question were reviewed by three raters who grouped the responses into 1. Professional 2. Semi-professional 3. Skilled 4. Unskilled 5. Student 6. Unemployed outside of the home Determined Reliability = none, since one-item scale Number of Items = 1 Item Number: 3 OUTSIDE INTERESTS -- this variable related to activities outside of the woman's home. Devised Item (a) Outside lnterests--this was a one—item scale lhave many interests A B C D E lhave no interests A B C D E Determined Reliability = none, since one item scale Number of Items = 1 Item Number: 10 LEVEL OF CONFLICT —- this variable related to (I) (2) (a) areas of conflict between the husband and wife, as well as feelings of the wife toward the husband. More specifically, it related to marital concensus, affectional expression, satisfaction and cohesion within the marriage. Devised ltems Regarding Conflict Dyadic Adjustment Scale Conflict (Devised ltems)--this scale measured conflict within the marriage by assessing the wife's feelings toward her husband as well as agreement/disagreement in other issues of importance to the couple. Determined Reliability = .94 Number of Items = 11 Item Numbers: 173, 174, 176, 177, 178, 179, 217, 218, 219, 220, 221 LIZ (b) Dyadic Adjustment Scale: (1) (2) (3) (ll) Djadic Concensus--items focus agreement in decision making and choice of friends, as well as agreement in couple's philosophy of life (Spanier, 1976, p. 24). Established Reliability = .90 Determined Reliability = .96 Number of Items = 14 Item Numbers: 184, 185, 186, 188, 190 191, 192, 193, 194, 195, 196, 197, 198, 199 Affectional Expression--this subscale dealt with sexual satisfaction and expression of physical affection within the relationship (Spanier, 1976, p. 24). Established Reliability = .73 Determined Reliability = .85 Number of Items = 4 Item Numbers: 187, 189, 213, 214a Dyadic Satisfaction--dyadic satisfaction measures general satisfaction within the relationship (Spanier, 1976, p. 24) These items included consideration of ending the relationship and amount of time the individual and her mate quarrel and get on each other's nerves. Established Reliability - .94 Determined Reliability = .83 Number of Items = 10 Item Numbers: 200, 201, 202, 203, 204, 205, 206, 207, 214b, 215 Dyadic Cohesion--included in this sub- scale were items regarding shared interest, working together on projects, and whether or not the couple had a stimulating exchange of ideas (Spanier, 1976, p. 24). Established Reliability: .86 Determined Reliability = .90 Number of Items = 5 Item Numbers: 208, 209, 210, 211, 212 l. Variable: Measure: Scale: J. Variable: Measure: Scale: 43 SUBMISSION/DOMINANCE -- this variable related to the general feeling of assertiveness on the part of the woman Devised ltems (a) Submission/Dominance--this scale measured the general level of assertiveness the woman felt best characterized her. More specifically, a set of bipolar items measured brave/ cowardly, assertive/unassertive, passive/ aggressive. It also explored the decision making balance or imbalance between the couple. .72 (Item 135 included) .76 (Item 135 excluded) Number of Items = 10 (9 with Item 135 excluded) Item Numbers: 133, 135, 157, 158, 159, 164, 180, 181, 182, 183 Determined Reliability LEVEL OF SELF-ESTEEM -- this scale related to the woman's feelings, or lack of feelings, or self worth Devised ltems (a) Self-Esteem--this scale measured the woman's feelings of self worth by using bipolar items to look at such characteristics as worthwhile versus worthless, useful versus useless, adequate versus inadequate, smart versus stupid, attractive versus unattractive, etc. Determined Reliability .77 (Item 166 included) .83 (Item 166 excluded) Number of Items = 10 (9 with Item 166 excluded) Item Numbers: 137, 144, 145, 146, 148, 149, 160, 161, 162, 166 K. Variable: Measure: Scale: 44 ANXIETY --this variable related to the woman's general nervousness and excitability. Devised Items (a) Anxiety--this scale measured the woman's ee ing of calm versus agitation as well as such physiological characteristics as sweaty palms and good versus poor memory. Determined Reliability = .58 Number of Items = 6 Item Numbers: 123, 130, 132, 146, 155, 156. CHAPTER IV DESIGN AND ANALYSIS Overview A questionnaire measuring the general areas of parental attitudes and attributes, sex role acceptance, marital conflict, and self-concept was given to women who received help while living for a short period of time in a battered women's shelter in the Ann Arbor, Michigan area. The questionnaire was also administered to a control group of women with comparable socioeconomic status who attended classes at the YWCA in Lansing, Michigan. A multivariate analysis of variance was used to compare the data obtained from these two g roups . Population Subjects were selected from a population of battered women in Washtenaw County. Washtenaw County has an operating shelter for battered women and at the time of this study, lngham County was in the process of developing a shelter. The control subjects were selected from a population of women attending "self—enrichment" classes at the YWCA in lngham County. This sample was selected because of the assumption that battered women who sought out the shelter were doing something for themselves just as were women who chose to take the self-enrichment classes. These two counties are 45 46 similar in that they both have large rural areas as well as major universities. Both areas also have industrial bases. Sample The sample used for this study consisted of twenty-four battered women from a shelter home for battered women in Washtenaw County and a control group of twenty-three non-battered women from self-enrichment classes at the Lansing YWCA. Demographic data about these groups can be found in Table 4.1. The two groups were similar in age, education, number of children and number of women employed outside of the home. However, the non-battered women who worked had slightly higher job skills than did battered women. This was an unexpected difference. The age range for battered women was 18 to 53 with a mean age of 30.79 years, and for non-battered women the ages ranged from 18 to 53 with a mean fo 29.46 years. The number of children ranged from 0 to 4 for battered women with a mean of 1.75 children, and for non-battered women the number of children ranged from 0 to 4 with a mean of 1.33 children. Years of education for the battered women ranged from 8 to 15 years with a mean of 11.79 years and for non-battered women the range was 9 to 17 years with a mean of 12.79 years. The majority of women in both groups were unemployed outside of the home but both groups included women who were skilled and unskilled laborers and semi-professional women. In terms of marital status, for battered women 11 of the women were married and separated but still in contact with their mates, 7 were married, 4 were divorced TABLE 4.1.--Summary of the Characteristics of the Sample: Battered Women (BW) and Non-Battered Women (N-BW). Battered Women (N=24) Non-Battered Women (N=23) Age: Range 18-53 Mean 30.79 Number of Children: Range 0-4 Mean 1.75 Number of Years of School Completed: Range 8-15 Mean 11.79 Employment Categories: Professional 0 (0)* Semi-Professional 2 (8.3) Skilled 2 (8.3) Unskilled 5 (20.8) Student 1 (4.2) Unemployed Outside Home 14 (58.3) Marital Status: Single -- Engaged -- Married 7 (29.17) Separated** 11 (45.83) Divorced 4 (16.66) Widowed -— Living with Intimate Partner 2 (8.33) 18-53 29.46 9-17 12.79 (8.7) (13) (13) (4.3) (4.3) (56.5) w-n-awww 20 (86.9) 3 (13.05) * Percentages are in parentheses. * *It is possible that BW answered "Separated" since they were living at the Shelter when this was filled out. If this is true and if separated implies being married but living apart 75% of BW can be considered married . 48 and still living with their mates, and 2 were single living with a man. All of the battered women had been abused by their mates. The non-battered women were all married or living with men and none had been abused by their mates. Procedure Following several pilot studies used to improve the question- naire's content and format, two sites were selected for the collection of data, the Washtenaw County shelter for battered women and the Lansing YWCA. Each program director was contacted, told the purpose of the study, and support for the study was secured. Seventy-five questionnaires were given to the Director of the Washtenaw County shelter for battered women, who was asked to meet with her staff and instruct them to distribute the questionnaires to the women within forty-eight hours after arrival at the shelter. It was anticipated that within a forty-eight hour time period1 the battered woman would reduce her emotional upheaval as a result of the beating and the move to the shelter. The battered women who volunteered were given several days to fill out the questionnaires without a strict time requirement. The total number of questionnaires were collected over a four-month time period resulting in twenty- four returns. During the same time period, seventy-five questionnaires were given to the Program Director of the Lansing, Michigan YWCA. 1Forty--eight hours was arrived at after speaking with . several counsellors who had worked with battered women. 49 The Director was instructed to give the questionnaires to women attending self-enrichment courses at the YWCA. The questionnaires were distributed to seventy-five women during the first week of class who were asked to return them within one week. The data was collected over a four-month time period with twenty-three returns. This group comprised the control group for the study. Testable Hypotheses This study answered three general questions about the per- sonality characteristics of the battered women. Since the three research questions were stated in generalized form and were not directly testable, the specific hypotheses were defined and are delineated below. Here, each research question is stated, followed by the research hypotheses which were constructed to provide at least a partial basis for answering each question. Since the hypotheses are stated in the null form failure to confirm these specific hypotheses is believed to be necessary if not entirely sufficient to answer the research questions. I. Sex Role Acceptance Research Question 1 : Are battered women oriented to more traditional sex roles than non-battered women? Null Hypothesis 1a: No difference will be found between battered women and non-battered women in their acceptance of the feminine sex role as measured by the BSRI. 50 Alternate Hypothesis 1a: The mean score for battered women on the femininity scale of the BSRI will exceed that of non-battered women. Null Hypothesis 1b: No difference will be found between battered women and non-battered women in their acceptance of the masculine sex role as measured by the BSRI. Alternate Hypothesis 1b: The mean score for battered women on the masculinity scale of the BSRI will be less than that of non— battered women. Null ijothesis 1c: No difference will be found between battered women and non-battered women in their acceptance of an androgynous sex role as measured by the BSRI. Alternate Hypothesis 1c: The mean score for battered women on the androgyny scale of the BSRI will be less than that of non- batttered women. ll . L'ife Histories Research Question 2: Do battered women differ in their life histories from non—battered women? Null Hypothesis 2a: No difference will be found between battered women and non-battered women with respect to feeling closer to their fathers than to their mothers. Alternate Hypothesis 2a: The mean score for battered women on the Parental Attitudes Scale for closest parent being the father will exceed that for non-battered women. 51 Null Hypothesis 2b: No difference will be found between battered women and non-battered women in having experienced violence in their childhood home. Alternate Hypothesis 2b: The mean score for battered women on questions about experience of childhood violence will exceed that for non-battered women. Null Hypothesis 2c: No difference will be found between battered women and non-battered women with respect to having mothers who had traditional female sex role preferences. Alternate Hypothesis 2c: The mean score for battered women will exceed that for non-battered women with respect to their mothers having had traditional female sex role preferences. Null Hypothesis 2d: No difference will be found between battered women and non-battered women with respect to having fathers who had traditional male sex role preferences. Alternate Hypothesis 2d: The mean score for battered women will exceed that for non-battered women with respect to their fathers having had traditional make sex role preferences. Null Hypothesis 2e: No difference will be found between battered women and non-battered women with respect to viewing their parents as exhibiting greater sex role enforcement. Alternate Hypothesis 2e: The mean score for battered women will exceed that for non-battered women with respect to their parents exhibiting greater sex role enforcement. 52 Null Hypothesis 2f: No difference will be found between battered women and non-battered women with respect to having experienced family protectiveness. Alternative Hypothesis 2f: The mean score for battered women will be less than that for non-battered women with respect to the measure of family protectiveness. Null Hypothesis Zg: No diffference will be found between battered women and non—battered women in their experience of family harmony and mother supportiveness. Alternate Hypothesis Zg: The mean score for battered women will be less than that for non-battered women with respect to female family harmony and mother supportiveness. Null Hypothesis 2h: No difference will be found between battered women and non-battered women in their experience of mother democracy and father democracy. Alternate Hypothesis 2h: The mean score for battered women will be less than that for non-battered women with respect to the measures of mother democracy and father democracy. I I I . Personal Adjustment Research Question 3: Do battered women differ from non-battered women in their personal adjustment? Null Hypothesis 3a: No difference will be found between battered women and non-battered women with respect to reporting of sexual difficulties and dissatisfaction. 53 Alternate Hypothesis 3a: The mean score for battered women will exceed that for non-battered women in reporting sexual difficulties and dissatisfaction. Null Hypothesis 3b: No difference will be found between battered women and non-battered women with respect to dependency. Alternate Hypothesis 3b: The mean score for battered women will exceed that for non-battered women with respect to dependency. Null Hypothesis 3c: No difference will be found between battered women and non-battered women with respect to feelings of incompetency outside of the home. Alternate Hypothesis 3c: The mean score for battered women will exceed that for non-battered women with respect to feelings of incompetency outside of the home. Null Hypothesis 3d: No difference will be found between battered women and non-battered women with respect to perceiving themselves as deserving of punishment. Alternate Hypothesis 3d: The mean score for battered women will exceed that for non-battered women with respect to perceiving themselves as deserving of punishment. Null Hypothesis 3e: No difference will be found between battered women and non-battered women with respect to social and emotional isolation . 54 Alternate Hypothesis 3e: The mean score for battered women will exceed that for non-battered women with respect to social and emotional isolation. Null Hypothesis 3f: No difference will be found between battered women and non—battered women with respect to being unemployed outside of the home. Alternative Hypothesis 3f: The mean score for battered women will exceed that for non-battered women with respect to being unemployed outside of the home. Null Hypothesis 3g: No difference will be found between battered women and non-battered women with respect to the number of outside interests. Alternate Hypothesis 3g: The mean score for battered women will be less than that for non-battered women with respect to the number of outside interests. Null Hypothesis 3h: No difference will be found between battered women and non-battered women with respect to the level of conflict with their husbands. Alternate Hypothesis 3h: The mean score for battered women will exceed that for non-battered women with respect to conflict with their husbands. Null Hypothesis 3i: No difference will be found between battered women and non—battered women with respect to submissiveness. 55 Alternative Hypothesis 3i: The mean score for battered women will exceed that for non-battered women with respect to submissiveness. Null Hypothesis 3j: No difference will be found between battered women and non-battered women with respect to level of self-esteem. Alternative Hypothesis 3j: The mean score for battered women will be less than that for non-battered women with respect to level of self-esteem. Null Hypothesis 3k: No difference will be found between battered women and non-battered women with respect to level of anxiety. Alternative Hypothesis 3k: The mean score for battered women will exceed that for non-battered women with respect to level of anxiety. Analysis The design used for this study was the Static-Group Compari- son in which women who have been beaten by their mates were being compared with a similar group of women who had not been physically battered. A questionnaire was administered to each woman in the sample. This design controlled for all internal sources of validity with the exceptions of such things as selection, mortality and interaction of selection and maturation (Campbell 8 Stanley, 1963, p. 8). Selection was on a voluntary basis with the women volunteering to fill out the questionnaire. The women in both the 56 battered and the non-battered groups were similar with respect to marital status, income, age, and educational level. Mortality was not a critical issue in this study, since there was only one observation made and only one instrument administered at that time. The null hypotheses, which state that there would be no difference between the group of battered women and the control group of non-battered women, were tested by analysis of variance. Reliability checks were completed in order to evaluate question groupings. In addition, frequency counts were completed on several items. Means were established for all scales. CHAPTER V RESULTS The primary method of analysis employed in this study was a multivariate analysis of variance (MANOVA). The MANOVA allowed for grouping of the dependent variables into three major questions and afforded the opportunity to determine the statistical significance of each question. A univariate analysis of variance was done in order to determine the significance of the minor questions that were represented by the dependent variables when the MANOVA was significant. The MANOVA utilized an alpha level of .01 in order to reject the null hypothesis. As a result of the large number of variables in this study a .01 alpha level was used to reduce the probability of getting significance from chance alone. Means were calculated in order to examine group mean differences for several items. Frequency counts were performed on items to see the results for single independent item variables. In order to best present the results, the significance of F for the MANOVA will be given by section. Each section will be pre- sented as a general hypothesis in the null form. When p < .01 the significant variables will be presented. If the section is found to be significant those variables which are significant will be presented. For those cases where the MANOVA indicates no significant differences 57 58 between groups for the section, the variables which are significant based on univariate analysis will be discussed in Appendix B. Section 1: Sex Role Acceptance General Hypothesis No difference will be found between battered women and non- battered women in their sex role acceptance as measured by the scales of the BSRI. There were no significant differences found between groups for this general hypothesis (Table 5.1). It was expected that a significant difference between groups would be found, particularly for the femininity subscale. However, this was not found by either the MANOVA or the univariate analysis. The mean score for non-battered women was greater than that for battered women but was not significantly higher on the masculinity and femininity subscales (Table 5.1). Based on the data collected there is failure to reject the null hypothesis. TABLE 5.1.--Multivariate Analysis of Variance—-Section 1: Sex Role Acceptance. F = 1.34 p < .27 Means Univariate BW NBW F p Dependent Variable (N=24) (N=23) Femininity 4.33 4.83 3.19 .08 Masculinity 4. 98 5.39 3.04 .09 Note: Significance is considered at an alpha level of .01. 59 With respect to androgyny, the difference between the feminity and masculinity score, neither group differs significantly from zero even though non-battered women were close to zero (Table 5.2). This leads to the conclusion that neither the battered group nor the non-battered group is highly sex-typed, and further supports the failure to reject the null hypothesis. Section 2: Life Histories The data for Section 2 hypotheses were analyzed by the use of the MANOVA with all scales included, and then again with the scales measuring Father Femininity and Father Masculinity deleted. These two scales were delted in the second analysis in order to determine the level of significance without those scales, since they contained large amounts of missing data. Gene ral Hypothesis No difference will be found between battered women and non-battered women with respect to life histories. The statistical significance at an alpha level of .01 was not obtained by the MANOVA either with or without the two scales measuring Father Femininity and Father Masculinity, and there was thus failure to reject the null hypothesis (Table 5.3. For a discussion of trends which appeared in the univariate analysis see Appendix B. 60 as. v a .3 u :5 .8; M 3:: .Nm.~ >9 mcoum mucocflto >c>mocpc< 2.: 9:22:35 >9 o:_m> ozmciu or: mum—EXocQQm :8 etc cam .33. u .: .muzcmE 2.2.313 ocm :rEEEon. Ucm >t£3umm<$ 88?: 95 9: .>:mu_.._QEm_ .Emm >9 mcop 59539. c_ mucopgm Cm Co oEEmm o>BmEcoc UoEnEou Eot >__mu_.:QEm “66250 8% 3.7 u 2.va 8a.; mm.~ 3a .. mmé macaw 83cm $17 .1. SWS $91. $4 and I 8.: 262 N 0.320 5.7 u 5.73 39¢ ~m.~ was - mm... 3m _ 0.320 >:c__:uwm.2 >:c_c_Emn_ 138m i *ocoom oucocflto x >939. pc< >c>mocpc< pcmpcmwm «Zoom cams. .25sz coEoS pocofimmicoz ocm .359 coEoB pocofimm cob. mmcoom >c>m0cpc< QDOLU Co co_um_:u_mui.~.m m4m<._. 61 TABLE 5.3.--Multivariate Analysis of Variance---Section 2: Life Histories. Dependent Variable F=.76; p<.08 F=1.01; p<.48 Means Scales Means Scales Included Deleted BW NBW BW NBW (N=24) (N=23) Closest Parent Father Positivity Mother Positivity Childhood Violence Father Abused Mother Mother Femininity Mother Masculinity Father Femininity Father Masculinity Sex Role Endorsement Family Protectiveness Female Family Harmony and Mother Supportiveness Mother Democracy Father Democracy 3 30 3 31 3.24 3 31 2 76 3.24 2 88 3 24 2.93 3.65 2.97 3.64 1.98 1.58 1.93 1.55 1.76 1.63 1.72 1.60 3.70 4.10 3.69 4.12 ‘3.23 3.22 3.19 3.19 2.84 3.07 3.41 3.77 3.00 2.81 3.03 2.87 3.33 3.72 3.30 3.70 2.80 3.62 2.89 3.59 2 48 3 39 2.60 3 37 2 79 3.17 2 83 3 15 Note: Significance is considered at an alpha level of .01 62 Section 3: Personal Adjustment The data for Section 3 was analyzed by use of the MANOVA wil all items included and then again with nine items deleted from various scales. Those items deleted were item 9 from the Dependency scale, items 124, 134, and 143 from the PerSonal Respect scale, item 15 from the Guilt scale, items 219 and 221 from the Conflict scale, item 135 from the Submission/Dominance scale, and item 166 from the Self Esteem scale. (See questionnaire in Appendix A.) The items were deleted because they were found to lower the reliabilities of the scales . General Hypothesis: No difference will be found between battered women and non- battered women with respect to personal adjustment. The MANOVA was run on fourteen dependent variables which included the following: Sexual Satisfaction, Affectional Expression, Dependency, Incompetency, Personal Respect, Guilt, Social/Emotional Isolation, Conflict, Concensus, Satisfaction, Cohesion, Submission] Dominance, Self-Esteem, and Anxiety. Statistical significance at an alpha level of .01 was obtained for the MANOVA for both complete- item and deleted-item variables and the null hypothesis was rejected (Table 5.4). In addition to the MANOVA a univariate analysis was per-— formed to explore the significance of each of the fourteen variables (Table 5.4). The univariate analysis yielded nine variables which were significant at an alpha level of .01. 63 ANN u 2V cmEoz, Dmcmwummicoz .2 u 25 5:53 8.65am .5. v a 5 ESEcmfi .2, .5. v Q Hm 585.55 cm 85. 3mg: 25 £5 ~55. 3.8 m; .55 352322 =5. 55. $5 $5 :5. 8. 55 a: 53mm :8 :5. 35.: $5 3.5 :5. 55...: .55 25 «8855985253 :5. 5m .5 8 .m 2 .N .. c2858 :5. 5 .2 S .m 8 .N Ecrufimzmm :5. 3 .5 K .m 5 .~ 8:38:50 :5. 5:: mm; 9.1m :5. :25 N5; 85 «65:8 :5. 2.: 5.~ 55 «8:28. 5:255:38 m8. 5: R5 55 5mm. 5.. 55 55 :50 5mm. .5. 35 as...” :5. S. 3.». 2; 888m 5:8ch 8»: mm. 2 .m mm .N 23858:. man. 2:. 5:” 55 m5. 2; 5...... 25 $866889 :5. 2.: 55m 35 icofimmcaxm 58:6ch :5. 2.5 $5 35 226333 53% a n. 52 gm 5 u_ 52 am mint; Cam—2 C092 6828 mam: :5. Va 55”“. nonstop: mEo: 80. v Q hNé H “— .ucoEpmzzoxx .8331 "m cofiuomioucmtm> Co m_m>_mc< 3m_..m>_:32i.=.m m4mocosqocn. 3303.6. HE2202, ommmhhcomoo 8. mc_£oEom EU 932m _ .: o>comop 3 mESoEom EU cofio _ A3 .: daemon 3 mchmEom EU >__mco_mmuuo _ E A: o>commp 8. mc_£>cm ED 5033 _ An: .: daemon 3 @550: EU _ Am. ”panama: >5 >9 pomanm co pwficmcfle >:mu_m>:a mm; _ can; .m?» t .m: *6: cozmosclmucaou >ucosqocmi.m.m mum<._. 67 0030030 25 cozmcm 50: EU c0503 33:31:02 £ mm.m cams. 5.5 5.5.. 5.5 5.5 5.5 5.5N 602:8ch N z N N N 5 55:09.... 2:582 ”2020; 005520. 5 3:033: $5 banana ..0 00.055sz 0.5; 30> :02; 00202: 9.35.5 0.03 coco 30.: ”0pr mm.m €005. 5.2 E. 5.5 m... m... m... «08:88.. 2 N F F p — N65539:. 33.0mm? u2m:>_0>.. owmmbhucm300cn. 33.0.55: u2955.5 memthm 205$ 33 03300 .5 00.05.050.58 0cm; 30> :05; ©0202: mExELU mm; coto 30: "EN— omN._ 500.2 0 Nu m .5 mp mamacoucod 5. _ N 5:302“. 33.552 H2950.5 005550-202 3.5 cams. mm . S N .5 N .3 own—coucmm 5_ 5 5 p 5 2 58:08“. 3282 ”2530.5 Smut/B 2050:. 33 000390 .5 00.95052: 30; 30> :05; 0030?; 95.5.6 mm; c0550 30: ”EN— 0mc00moa 02 5302 EoEom 2.9.0.3qu C030 coco 30> omcoamom . —~— £25030 ”3:300 >uco3qmculdd w..m<._. 68 5 .fi 2 N .55 f 3:835 oz .5.» N u n .5050 x m .5 F N .5 5 ..0ch5m ._ 5.5 P - - 3.5.5.. :33: 355902 .-_ .. I N... F 3.05 :33: 0005902 .5 - - - - _.._0 82902 5 m .5 F - - >00 805902 .0 m .5 P N .5 5 L050“. . 0 - - - - 53mm .0 u .. N. .m— 5 0050.5 .0 n u .. u A0_mE0: .._03_m>nmm .9 x I N.5 — A0225 03:93.5 .0 33 0030.08 >__mDX0m mm; _ 0:50 0 m< .35 0955:0800 0530.355 0mmuc00000 0N:_0mn< c0803 .0000fimmncoz C0803 U000fimm .35 0030030 "3:300 >0:0300¢as.n.m mjm<._. 69 women responded and there were no significant clusters in the responses (Table 5.7). Conclusion Based on the frequency counts done on the above items, it was interesting to note that a large number of the battered women felt that they deserved to be physically abused by their mates. Although this study does not confirm that alcohol is a problem in the abuse received from their mates, it was interesting to note that the majority of women who responded to this question with respect to parental abuse felt that alcohol had been a contribut- ing factor. 212m The dependent variables under consideration in this study were evaluated in three sections: (1) Sex Role Acceptance, (2) Life Histories, and (3) Personal Adjustment. These sections were analyzed by a multivariate analysis of variance. When the MANOVA was significant a univariate analysis of variance was used to evaluate the significance of each variable. Frequency counts were run on five items in order to explore the absolute frequency of response for the battered women. The results of the multivariate analysis of variance were insignificant for sex role acceptance (p. < .27) and life histories (p < .684 all scales included and p. < .477 with two scales deleted) resulting in failure to reject the null hypothesis at an alpha level of .01. The results of the multivariate analysis of variance for 70 personal adjustment were significant (p < .001) at an alpha level of .01, and the null hypothesis was rejected. Two univariate analyses of variance were performed for the personal adjustment variables, the first containing all items and the second deleting those items which had been found to lower the reliabilities of the scales. The significant variables at an alpha level of .01 were sexual satisfaction, affectional expression, marital con- census, marital satisfaction and marital cohesion. The means on all were higher for non-battered women. Dependency was significantly higher for on-battered women at a p < .05 alpha level with all items included. Battered women had significantly higher means on sociall emotional isolation, marital conflict, submission/dominance, and anxiety. Based on this data it appears that non-battered women have a higher level of personal adjustment. Frequency counts were done on items 118, 121a-c, and 167, in order to evaluate the responses of the battered women. The majority of the battered women felt that they had been deserving of the abuse they received. Drinking was frequently involved in the abuse the women received from their mothers and fathers but ll5.8 percent said that it was never involved in the abuse from their husbands. Of those battered women who had been sexually abused as children (33.33% of the total sample), the largest number had been sexually abused by their brothers. CHAPTER VI DISCUSSION In this chapter conclusions based on data analysis, methodo- logical problems encountered, and implications for future research will be explained. Additionally, alternative methods for overcoming the methodological problems will be suggested. General Methodological Problems with the Questionnaire The two major problems affecting the questionnaire were its length and its wording, and these problems may have played a part in the failure to obtain significance in sections 1 and 2 of this study. The length of the questionnaire made it a major undertaking to fill out. If a woman attempted to fill it out in one sitting she may have become fatigued by the time she reached the end. This fatigue factor may have led to carelessness and [or lack of interest in filling it out accurately. This may have been complicated by the emotional loading of the questions for the women. The wording of the questions was interpreted by the staff at the shelter as being "too middle class." It may have been, also, that the vocabulary was too complex for many women. These factors may have played a part in the failure of both the battered and the non-battered women to return many of the questionnaires. 71 72 Another problem indirectly related to the questionnaire was the testing situation. The battered women answered the questionnaire while staying at the shelter. Since the shelter is a safer place for the women they undoubtedly felt more in control there than they did while in the violent home situation. This control may have led the women to answer in a more positive way or in a more controlled manner. It is also possible that much of the anger experienced by the women had dissipated within 48 hours of being at the shelter. One must also consider the staff/client interaction as a possible influ- ence on the response of the women to the questionnaire. The staff tended to be very supportive of the women and this may have led to a heightened feeling of self-esteem and less passivity on the part of the battered women. These factors make it difficult to obtain an unbiased picture of the battered woman. The ideal setting would be to give the woman the questionnaire before she leaves the battering situation. This is difficult to do since gaining access to these women is problematic. A possible solution to this problem would be to dis- tribute the questionnaires to mental health therapists and physicians who see the battered women in treatment while they are still living in the home, and have them ask the women to fill them out. It is of interest that of the 75 questionnaires given out at the shelter 24 were returned and many of those not returned were found torn up and placed in cracks in walls, stuffed into the box springs of beds and thrown into wastebaskets. Others were found with Obscenities written on them. It appears as though the questionnaire elicited a great deal of anger from the women. 73 The anger which surfaced and was directed at the question— naire may have actually been hidden underneath the passivity demonstrated by the women in the battering situation. This idea gains support in the work done by Walker (1981): The passivity covers much of the anger that those women cannot express directly for fear of precipitating another battering incident. When they perceive it is safe to do so, this anger is usually expressed indirectly. Battered women learn that it is less dangerous to express anger in this indirect manner, and in continuing to do so, their angry feelings may generalize to others, including therapists (p. 86). If this is true for those women who destroy their question- naires, what does it mean for those women who were compliant and filled them out? It is very likely that this anger may have led to a source of bias in response to many of the questions. Walker (1981, p. 86) goes on to say that ". . . they may appear to passively comply with a request and sabotage it later on." Preventing this source of bias from occurring is a difficult task. It may prove useful to have the shelter staff work with the women in order to elicit some of the anger previous to filling out the questionnaire. Another possibility would be to include within the questionnaire an instrument that would allow a measure of anger that could be used as a weight for the other scales. It is important that future research done with a shelter population be aware of, and give consideration to, this hidden anger. Discussion of Hypothetical Variables This section will be presented in three parts, one for each of the general hypotheses that was investigated in this study. Each part will be divided into findings and methodological problems. 74 Sex Role Acceptance Findings: There were no significant differences found between battered women and non-battered women on the masculinity or femi- ninity subscales of the Bem Sex Role Inventory. The androgyny measure from the BSRI also indicated that neither group of women was highly sex-typed. This finding is of interest since a large amount of literature in the area has pointed to sex role socialization as a major contributing factor to acceptance of battering. There are theoretical reasons for the failure to find significant differences between battered and non-battered women in the area of sex role acceptance. It had been hypothesized that battered women would be more stereotypically female than non-battered women. There has been reference made in the literature that the battered women are often amibitious, well-educated and function well outside of the home, yet within their marriages they resort to stereotypically traditional behaviors (Rounsaville, 1978, p. 19; Walker, 1978, p. 529). What this implies for this particular study is that women who are living in a shelter situation are outside of the home and may see themselves as more assertive and less passive. If, in fact, these women are now seeing themselves in a more positive light they may respond to the BSRI as they now perceive themselves. Another possibility is that the battered woman was not con- sistently a stereotypically traditional woman and that this was one factor which led to the beating. In a study by Rounsaville (1978) little evidence was found for men assuming any traditionally female tasks but the women performed many traditionally masculine functions 75 and felt themselves to be at least as capable as men in carrying out these functions (pp. 26-27). The literature indicates that violence occurs between people who are in superior-subordinate roles. Violence may occur when the woman who is in the subordinate role moves out of that position. Support for this idea comes from Rounsaville (1978): A particularly volatile combination seems to be a jealous, possessive man with paranoid tendencies and a counter- dependent, indomitable, passive—aggressive woman. If the man was less dependent, he could tolerate her inde- pendent strivings. If the woman was less dependent, she would be able to leave more easily. Alternatively, if she was more submissive and willing to be dominated, she might be less likely to be abused (p. 22). Additionally, if the woman is more skilled verbally and attains the upper hand in an argument the man may resort to physical violence in order to regain the superior position. These two theoretical reasons for the lack of support for the hypothesis of sex role acceptance are not mutually exclusive. It is very possible that upon entering the shelter the woman's awareness of her skills and strengths increases. If she had been conflicted earlier about her "masculine" competencies she may, through the sup- port of other women in the shelter, become less conflicted. It is also possible that by merely leaving the violent home she gains a sense of increased competency. Methodolggical Problems: It is possible that methodological problems, either alone or in conjunction with the theoretical explana- tions, are responsible for the failure to reject the null hypothesis pertaining to sex role acceptance. The BSRI was the last part of 76 the questionnaire and after answering in excess of 200 questions the women may have been too fatigued, or bored, to answer accurately. It may have been a more reliable measure if placed in the middle of the questionnaire. In addition, the terms used in the BSRI may have been too complex for some of the women to understand. The BSRI may not be a valid measure of sex role acceptance for women with average or below average levels of education. Many of the terms used are not commonly used in everyday conversation, e.g., con- scientious, flatterable, analytic, gullible, individualistic, assertive. A solution to this problem may have been to simplify the terms by using words that are less complex but of equivalent meaning. Life Histories Findings: There were no significant differences found between battered women and non-battered women with respect to the general hypothesis dealing with life histories. This finding is of interest since there has been much debate regarding the role early childhood experiences play in the battered woman syndrome. It is possible that the major effect of parenting in the battering situation is in the life of the batterer and not in the life of the battered women. This would explain the lack of significant findings between the battered women and the non-battered women with respect to life histories. The idea that early family violence may be a major influence on the batterer gains support in the study by Prescott and Letko (Roy, p. 77): 77 What is suggested, however, is that some aspects of early relationships and some early forms of social learning may further be reinforced by traditional sex role upbringing, and may facilitate the expression of violence in marriage. Some men have tended to be socialized in a manner that condones aggres- sion, and this socialization when combined with an early childhood experi- ence of violence may facilitate the expression of violence in the marriage. Women have been socialized in a manner that condones passivity, and this socialization, when combined with a violence-prone male, has dangerous potential. See Appendix B for alternative discussion of trends in the data. Methodological Problems: A methodological problem that is specific to the section on life histories and may have contributed to the lack of significant findings is the period in time that the woman answered the questionnaire. For both the battered and non-battered women there was a fairly large time lapse from when they were living at home with their parents. The mean ages for battered women was 30.79 years and for non-battered women it was 29.46 years. This may indicate that they have not lived with their families of origin for approximately ten years. If this is true the results may be biased by inaccurate recall. Additionally, the stress of battering situation at the time the questionnaire was filled out may have also influenced the response for the battered women. These results must also be viewed with some caution as a result of the small sample size. The sample size for both battered and nonbattered women was less than twenty-five per group, which affects the power of the test. This is a problem common to much of the research done in domestic abuse since obtaining a large sample 78 size is a difficult task. It is critical that more research in this area be done and that more data be gathered in the area of life histories with greater sample sizes. Additionally, one must be aware that the reliabilities on several scales were less than adequate as they fell below the .70 level. It is imperative that the area of life histories be studied with a more reliable scale for the population of battered women. Although most of these scales were devel0ped by Spence and Helmreich and found to be adequate for their use, it is very likely that their sample was significantly different from the sample in this study. Personal Adjustment Findings: The following conclusions were obtained from analysis of the variables for Personal Adjustment: 1. Battered women experienced more social and emotional isolation than non-battered women. 2. The battered women were found to be more submissive than the non-battered women. 3. Anxiety was significantly greater for battered women than for non-battered women. ll. Battered women felt deserving of their punishment. 5. Battered women experienced more conflict in their marriages than did non-battered women. 6. Non-battered women experienced more sexual satis- faction in their relationships than did battered women. 7. The relationships of the non-battered women contain more expression of physical affection. 8. There was significantly more husband-wife concensus in the non-battering relationship. This can be viewed as more agreement in decision making and choice of friends, as well as in the couple's philosophy of life. 79 9. The non-battered women expressed significantly greater satisfaction in general within the relationship. 10. Cohesion within the marriage was significantly greater for non-battered women than for battered women. Although non-significant results do not mean that no differ— ences between groups exist, since measurement and other factors may be influencing the results, it is interesting to note the following non- significant results: 1. There were no significant differences found between the groups on the dependency variable. 2. A feeling of incompetency/competency outside of the home did not differ significantly for the two groups. 3. No significant differences were found for personal respect. 4. Guilt did not differ significantly between the groups. 5. There were no significant differences found in the area of self-esteem. These results are summarized in Figure 6.1. Taking the non-significant results first, all of these areas are primarily representative of internal feelings of the woman about herself. The lack of significant differences in these areas suggests that the battered woman does not feel as bad when she is alone but rather in the context of a relationship. This idea gains some support in research by Walker (1978) who worked with a different sample of women: Battered Women seem to be most affected by feelings of helplessness in their relationships with men . . . . Many women are well-educated, ambitious, and function extremely well in high status positions. However, when it comes to their marriages or in other social relationships with men . they typically defer to the men to make decisions, even if they have manipulated the choices behind the scenes. Direct 80 communication is conspicuously absent from the battering relationships studied to date (Walker, 1978, p. 529). Rounsaville found that the women in his study were competent in relationships with children, with their families of origin and at work, but were impaired in leisure time activities and in relationships with their partners (Rounsaville, 1978, p. 19). This may also play a part in battering situations where the batterer feels that his posi- tion is threatened by a more skillful wife and dominance is maintained .by violence (Rounsaville, 1978, p. 26; Goodstein 8 Page, 1981, p.1038). MEANS SlGNlFlCANTLY MEANS SIGNIFICANTLY HIGHER FOR HIGHER FOR BATTEREDIWOMEN NON—BATTERED WOMEN Social/Emotional Isolation Sexual Satisfaction Conflict Affectional Expression Submission/Dominance Concensus Anxiety Satisfaction Cohesion NON-SIGNIFICANT VARIABLES Incompetency Personal Respect Guilt Self Esteem Figure 6.1.--Differences on Variables within Section 3: Personal Adjustment. 81 In comparing significant findings for the two groups (see Figure 6.1), the non-battered women were found to have more affec- tional expression in their relationship, more sexual satisfaction, more husband and wife concensus, more marital satisfaction, and more cohesion. Those results in and of themselves were not surprising since they speak to the quality of the non-battering relationship. Battered women, on the other hand, were found to have sig- nificantly more social and emotional isolation, more conflict, were more submissive, and had significantly more anxiety. These results are of interest since they point to two people in an unhealthy relationship with an anxiety base. This gains support in research by Walker (1980) on the three phase cycle of violence published subsequent to the beginning of this study. The three phases are (l) tension building phase, (II) accute battering phase, and (Ill) kindness and loving behavior phase (Walker, 1980, pp. 55-70; Walker, 1981, p. 82; Hilberman, 1980, p. 1339). It appears that the social and emotional isolation that the woman experiences plays a major role in her staying in the marriage. She believes that all marriages are abusive. In addition the isolation leaves her no outlets for increasing tension and she fears not only another assault but a constant struggle to control her own aggressive impulses (Hilberman, 1980, p. 1342). The sub- mission exhibited by the battered woman can be viewed as both an attempt to ward off the acute battering phase as well as an attempt to defend against her own homicidal rage (Hilberman, 1980, p. 1342; Walker, 1981, p. 86). The woman also fears that there is no way out 82 of the situation and assumes responsibility for keeping the environ- ment free from events which would trigger a battering (Walker, 1981, p. 85). The combination of isolation and submission as well as the buildup of the woman's own anger and fear lead to escalating anxiety. It is the unbearable build-up of tension and anxiety that leads to the acute battering phase. This battering phase serves as an unhealthy release for the tension felt by both the husband and wife. It is unclear what sets off the phase II violence but it is seldom the behavior of the woman. Frequently it is an external event or the internal state of the man, although one possibility is that the increase in tension makes it difficult for the woman to step back and allow a cooling off period (Walker, 1980, p. 60). The third phase of kindness and loving behavior seems to be the calm after the storm, and is looked forward to by both the batterer and his wife. This stage is marked by the batterer's promise that he will never do it again and that he is truly sorry (Walker, 1980, p. 65). It is at this point that the wife sees her husband as sick and feels that the only way he will get better is if she takes care of him (Walker, 1980, p. 67). Although there were no significant differences between battered women and non-battered women with respect to guilt, the majority of battered women felt that they deserved the beating. A strong possibility is that the beating may serve to absolve the woman of her guilt. When considering the three phase cycle of violence it is use- ful to look at the methods of communicating for both, battered and non-battered women. Husband and wife concensus and cohesion are 83 critical to a healthy relationship as are marital satisfaction and affectional expression. The battered women were significantly lower in all of these areas. What this may mean is that if concensus is seldom a possibility, tension and arguments will occur whenever it is necessary to make a decision. Since the couple is lacking in marital cohesion, which implies that they are unable to talk things through, and affectional expression is also missing, conflict resolution becomes an even bigger problem. If the woman attempts to talk to her husband and cohesion is not reached, the frustration will increase. As the frustration increases any alternatives for resolving differences diminish. If the woman becomes submissive and the rela- tionship still does not become cohesive she must let out her frustra- tion. The frustration may be released in a passive-aggressive way which will lead to the beating. If this is true, the battered woman may not be neurotic but may provoke the violence and the batterer becomes the "bad guy." This also may explain why the battered women were not significantly different from non-battered women in the area of guilt; being battered absolves them of their guilt. Another possibility for the significant differences that occurred between battered and non-battered women is simply that these differ- ences speak to the overall quality of the relationship. It had been anticipated that non-battered women would be involved in healthier relationships, especially where decision making was involved. The non-battered women did score higher in general satisfaction with the relationship, marital cohesion, marital concensus, affectional expres- sion and sexual satisfaction, all of which indicate a healthier 84 relationship than that of battered women. Since the questionnaire was given to the battered women shortly after they left the battering situation it is very possible that the abusive qualities of the relation- ship were highlighted in their minds and that their personal weak- nesses were overshadowed. Methodological Problems: Whether or not the non-significance of five variables reflects true similarities among battered and non- battered women, or whether methodological problems contribute to the non-significance is unclear. One possible methodological problem which may have contributed to the non-significance is the low reli- ability values. Several of these variables contained only a small number of items, i.e., three, and may have proven more powerful with more items. Additionally, it is possible that the raters who were used to categorize the items in the pilot study had a different experi- ential definition of dependency, competency and guilt. A possible solution to this problem would be the use of scales that have a fairly high, and previously tested, reliability. This does not, however, explain the low reliability and low alpha level obtained for personal respect since it was a part of a larger reliable measure. A possibility for the low reliability of per- sonal respect is that it does not stand up as well when removed from the larger scale. 85 Implications for Future Research The battered wife syndrome has only received much attention in the past decade and research has been sketchy. This leaves a field that is wide open for research especially in the area of psycho- logical factors involved in the battering relationship. Many of the methodological problems stated in this study need to be overcome even though the dynamics of the situation may make it difficult to do. Sex role acceptance needs to be studied in greater depth with methodological changes. One possibility would be to study the women while still living in the battering situation. A suggestion is to give the measure to physicians and mental health therapists who are work— ing with the battered women while they are still living in the home. It is important that they be given to the women close to the onset of the treatment. Another interesting method of studying sex role acceptance would be to have both individuals in the relationship fill out the scales as they perceive the other person and as they perceive themselves. This may shed light on the congruence between per- ceived and actual sex role acceptance. There is a need for more research in the area of families of origin. This study points to some interesting statistical trends with respect to the mothers of battered women being abusive of their daughters. It would be useful to study this in more depth with a large sample size. Additionally, it would prove useful to study the differences between the mothering and fathering received by the battered woman. Along with this it will be important to find a way of overcoming the bias resulting from recall over a lapse of years. The 86 idea of the battered woman as a borderline personality needs a great deal more study since it carries with it serious treatment implications. It would also be useful to explore differential treatment of siblings, from the family of origin, and whether or not they too are in battering situations. Along with this it would be of interest to explore whether or not in the families of origin of battered women there had been differential treatment of male children. When con- sidering families of origin further research in the area of learned helplessness and its onset would be useful. The whole area of locus of control, not only personal respect, has important implications for battered women and needs further investigation. It is also of importance to study in greater detail the areas of self esteem, competency, guilt, and dependency to determine whether or not the lack of significance in this study occurred for theoretical or methodological reasons. REFERENCES 87 REFERENCES Abramson, C. Spouse Abuse: An Annotated Bibliography. Pittsburgh: Chatham College, 1977. Ball, Margaret. Issues of Violence in Family Casework. Social Casework, January 1977, pp. 3-12. Ball, Patricia 0., 8 Wyman, Elizabeth. Battered Wives and Power- lessness: What Can Counselors Do? Victimology: An International Journal, 1978, 2, 5115-552. Barry, H.; Bacon, M. K.; 8 Child, I. L. A Cross-Cultural Survey of Some Sex Differences in Socialization. Journal of Abnormal and Social Psychology, 1957, 5_5_, 327-332. Bem, Sandra. The Measurement of Psychological Androgyny. Journal of Consulting and Clinical Psychol_ogy, 19711, 52, 155-152. Bem, Sandra. Sex Role Adaptability: One Consequence of Psycho- logical Androgyny. Journal of Personality and Social Psychology. 1975, a, 63u-6ll3. Brainard, P. A Follow-up Evaluation of the lngham County Prose- cuting Attorney's Domestic Abuse Program. Unpublished paper, December 1978. Broverman, Inge; Broverman, Donald; Clarkson, Frank; Rosenkrantz, Paul; 8 Vogel, Susan. Sex-Role Strategies and Clinical Judgments of Mental Health. Journal of Consulting and Clinical Psychology, 1970, 33, 1-7. Campbell, Donalt T., 8 Stanley, Julian C. Experimental and Quasi Experimental Designs for Research. Thicago: Rand McNally College Publishing Co., 1963. Coleman, Karen H., Weinman, Maxine L., 8 Hsi, Bartholomew. Factors Affecting Conjugal Violence. Journal of Psychology, 1980, 1_05_, 197-202. Flynn, John P. Recent Findings Related to Wife Abuse. Social Casework, 1977, g, 13-20. 88 89 Follingstad, Diane R. A Reconceptualization of Issues in the Treat- ment of Abused Women: A Case Study. Psychotherapy: Theory, Research and Practice, 1980, 11, 294—303. Gayford, J. J. Wife Battering: A Preliminary Survey of 100 Cases. British Medical Journal, 1975, 15, 1911-197. Gelles, Richard J. Abused Wives: Why Do They Stay? Journal of Marriage and the Family, 1976, 38, 659-668. Gelles, Richard J. No Place to Go: The Social Dynamics of Marital Violence. Battered Women. Edited by Maria Roy. New York: Van Nostrand Reinhold Co., 1977, 116-62. Gillman, Irene S. An Object-Relations Approach to the Phenomenon and Treatment of Battered Women. Psychiatry, 1980, {I}, 346-358. Goodstein, Richard K., 8 Page, Ann W. Battered Wife Syndrome: Overview of Dynamics and Treatment. American Journal of Psychiatry, 1981, 138(8), 1036-1044. Hanks, Susan E., 8 Rosenbaum, C. Peter. Battered Women: A Study of Women Who Live with Violent Alcohol-Abusing Men. American Journal of Orthopsychiatry, 1977, g, 291-306. Hilberman, Elaine. "Overview: The 'Wife-Beater's Wife' Reconsidered. American Journal of Psychiatry, 1980, £3101), 1336-13117. Horney, Karen. Feminine Pchhology. New York: W. W. Norton and Co., Inc., 1967. Martin, Del. Battered Wives. New York: Pocket Books, 1977. McShane, Claudette, compiler. Annotated Bibliography on Woman Batterin . Milwaukee: Center for Advanced Studies in Human Services, School of Social Welfare, University of Wisconsin, July 1977. O'Brien, John E. Violence in Divorce-Prone Families. Journal of Marriage and the Family, 1971, 33(4), 692-698. Prescott, Suzanne, 8 Letko, Carolyn. Battered Women: A Social Psychological Perspective. ln Battered Women. Edited by Maria Roy. New York: Van Nostrand Reinhold Co., 1977. Rounsaville, Bruce J. Battered Wives: Barriers to Identification and Treatment. Paper presented to the American Psychiatric Association Meetings, May 1977, Toronto. 90 Roy, Maria. A Current Survey of 150 Cases. In Battered Women. Edited by Maria Roy. New York: Van Nostrand Reinhold Co. , 1977. Schuyler, Marcella. Battered Wives: An Emerging Social Problem. Social Work, 1976, 21, l188-91. Sharness, Natalie. Psychological Aspects of Wifebattering. In Battered Women. Edited by Maria Roy. New York: Van Nostrand Reinhold Co., 1977. Spence, Janet Taylor, 8 Helmreich, Robert L. Masculinity and Femininity. Austin: University of Texas Press, 1978. Steinmetz, Suzanne K., 8 Straus, Murray. The Family as a Cradle of Violence. Society, 1973, 1_l_)_, 50-58. Walker, Lenore E. Battered Women and Learned Helplessness. Victimology: An International Journal, 1978, 2, 525-534. Walker, Lenore E. Battered Women: Sex Roles and Clinical Issues. Professional Psychology, 1981, _1_2_, 81-91. Walker, Lenore E. The Battered Woman. New York: Harper 8 Row, 1980. APPENDICES 91 APPENDIX A QUESTIONNAIRE 92 DIRECTIONS: This questionnaire has been devised to explore personal and interpersonal relationships that have been important influences on your life, and therefore it is a long questionnaire. You may not want to answer it in one sitting. Answer as much as feels comfortable in one sitting and then go back to it later. You will be able to finish it in one week. When finished, please return the questionnaire to the person who gave it to you. 93 Please fill in all blanks and circle the appropriate response. 1. Age: 2. Number of years of school completed: 3. Type of job you now hold (job title): 4. Number of children living with you: 5. Marital status: (circle one) a. single b. engaged c. married d. remarried e. separated f. divorced g. widowed h. living with intimate partner 6. If married, husband's age: 7. If married, husband's occupation: This set of questions ask you to describe yourself. Each item has a scale with the letters A, B, C, D and E with (A) being most like you and (E) being least like you. Please circle the letter that comes closest to how you see yourself. 8. I could be A B C D E I could never financially be financially independent independent 9. I would like to A B C D E I would not like be financially to be financially independent independent 10. I have many A B C D E I have no interests interests 9’4 RELIGION 11. 12. 13. Ill. 15. As an adult I am (circle one) Catholic Jewish Protestant Atheist Other A B C D E If you are Protestant, please state denomination: As an adult, I am (circle one) Very Somewhat Not Very Not at All Religious Religious Religious Religious Religious A B C D E As an adult, I go to church or synagogue (circle one) Very Often Often Occasionally Seldom Not at All A B C D E As an adult I feel guilty if I don't go to church or synagogue (circle one) Very Often Often Occasionally Seldom Not at All A B C D E 95 PARENTAL ATTITUDES l. The questions ask for information about your parents’ attitudes and actions. "Parent” includes step-parent, foster parent or any other adult guardian who had been responsible for you all or most of your life. If a question asks about "parents" and you were brought up by only one, answer for him or her. Answer every item by picking the letter on the scale below which best describes how characteristic or uncharacteristic it is as it applied to your experience in your family. Very Very Characteristic A B C D E Uncharacteristic A B C D E 16. Members of my family are very close and get along amazingly well. A B C D E 17. When I was little, my parents considered it their business to know what I was up to all the time. A B C D E 18. At home I had a quite definite daily schedule I was expected to follow. A B C D E 19. If I go/went on after I finish my education and have a very successful career, my parents will be/would have been very pleased. A B C D E 20. Relative to friends my age, there were fewer family rules and regulations l was expected to follow. A B C D E 21. If I have any children, I expect to bring them up very similarly to how I was brought up. A B C D E 22. Our family has always done lots of things together. A B C D E 23. My parents encouraged me to stick up for my rights and to fight back if anybody tried to push me around. 96 II. All of the questions in this section refer to your mother or other female guardian. If you grew up without a mother or female guardian, leave this section blank and go on to Section III, Question 48. Very Characteristic ABCDE 211. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Very A B C D E Uncharacteristic My mother believed there was no reason why she should have her own way all the time any more than I should have mine. My mother encouraged me to talk to her about my troubles. There were rules in my family but lots of times my mother didn't really care if I lived up to them. My mother didn't mind if I played with toys that were supposed to be for the opposite sex. When I did something I shouldn't, my mother tried to get me to understand why I was wrong rather than simply punishing me. My mother encouraged me to do my best on everything I did. My mother didn't want me to bother her with unimportant little problems. I received a good deal of physical affection from my mother. I would describe my mother as a strict parent. When I look back, I think my mother criticized me or punished me a lot more than I deserved. l was expected to do what my mother told me to with little discussion or explanation. My mother always has set high standards for me to meet. I was encouraged to tell my mother if I believed a family rule was unfair. 97 Very Very Characteristic A B C D E Uncharacteristic A B C D E 37. I feel that my mother has almost always approved of me and the things I do. A B C D E 38. My mother is very sympathetic to "women's lib." A B C D E 39. My mother frequently praised me for doing well. A B C D E 110. My mother tried to impress upon me that getting along with people was one of the most important things I could learn. A B C D E 111. My mother and I argued a lot about what I should be doing and how I should behave. A B C D E 112. My mother always took an interest in my activities. A B C D E 113. My mother frequently criticized what I was doing. A B C D E 119. My mother was always careful and cautious about what she'd let me do for fear I'd get hurt. A B C D E 115. My mother was so inconsistent in what she expected of me I just gave up trying to understand her. A B C D E 116. My mother believed I had a right to my own point of view and allowed me to express it. A B C D E 47. When I did something I wasn't supposed to and my mother found out about it, she very often let me get away with it. 98 Ill. All the questions in this section refer to your father or other If you grew up all or most of the time without your father or a male guardian in your home, leave this section blank and go on to Section IV, Question 72. male guardian . Very Characteristic ABCDE 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. Very A B C D E Uncharacteristic My father believed there was no reason why he should have his own way all the time any more than I should have mine. My father encouraged me to talk to him about my troubles. There were rules in my family but lots of times my father didn't really care if I lived up to them. My father didn’t mind if I played with toys that were supposed to be for the opposite sex. When I did something I shouldn't, my father tried to get me to understand why I was wrong rather than simply punishing me. My father encouraged me to do my best on everything I did. My father didn't want me to bother him with unimportant little problems. I received a good deal of physical affection from my father. I would describe my father as a strict parent. When I look back, I think my father criticized me or punished me a lot more than I deserved. l was expected to do what my father told me to with little discussion or explanation. My father always has set high standards for me to meet. I was encouraged to tell my father if I believed a family rule was unfair. 99 Very Characteristic 61. 62. 63. 6“. 65. 66. 67. 68. 69. 70. 71. Very A B C D E Uncharacteristic I feel that my father has almost always approved of me and the things I do. My father is very sympathetic to "women's lib." My father frequently praised me for doing well. My father tried to impress upon me that getting along with people was one of the most important things I could learn. My father and I argued a lot about what I should be doing or how I should behave. My father always took an interest in my activities. My father frequently criticized what I was doing. My father was always careful and cautious about what he'd let me do for fear I'd get hurt. My father was so inconsistent in what he expected of me I just gave up trying to understand him. My father believed I had a right to my own point of view and allowed me to express it. When I did something I wasn't supposed to and my father found out about it, he very often let me get away with it. 100 IV. 72. 73. 74. 75. 76. If you did not grow up with both your mother and father [or step— or foster parents ) all or most of the time, leave this section blank and go on to Section A, Question 77. When you had a problem, whom did you confide in? eaves My father almost always My father more often than my mother My father and mother equally My mother more often than my father My mother almost always. My mother and father have always agreed quite closely on how children should be brought up. eases While 90‘? d. e Very characteristic Often characteristic Only sometimes characteristic Often uncharacteristic Very uncharacteristic l was growling up, I felt: Much closer to my father than my mother Somewhat closer to my father than my mother Equally close to my mother and my father (or not close to either) Somewhat closer to my mother than my father Much closer to my mother than my father. My ideals are: 909 d. e Much more similar to my father’s than my mother’s Somewhat more similar to my father’s than my mother's Equally similar to both my parents (or not similar to either) Somewhat more similar to my mother's than my father's Much more similar to my mother's than my father’s My personality is: 9°? 90. Much more similar to my father's than my mother's Somewhat more similar to my father’s than my mother’s Equally similar to both my parents (or not similar to either) Somewhat more similar to my mother's than my father's Much more similar to my mother’s than my father’s 101 PARENTAL ATTRIBUTES A. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. MOTHER ’S A TTRIBUTES The following items refer to your perceptions of your mother. If you were brought up most of your life by a stepmother, foster mother, or other female guardian, please answer for her instead. If there was no woman regularly in your household who was responsible for your upbringing, go to Section B. Not at all independent Not at all emotional Very passive Not at all able to devote self com- pletely to others Very rough Not at all helpful to others Not at all competitive Not at all kind Not at all aware of feelings of others Can make decisions easily Gives up very easily Not at all self- confident Feels very inferior Not at all under- standing of others Very cold in relations with others Goes to pieces under pressure A A A A 102 WWWCD a: O 0000 UUUU l‘f'll'l'll'fll'l'l 0 Very independent Very emotional Very active Able to devote self completely to others Very gentle Very helpful to others Very competitive Very kind Very aware of feelings of others Has difficulty making decisions Never gives up easily Very self-confident Feels very superior Very understanding of others Very warm in relations with others Stands up well under pressure 11 PERSONAL DA TA 109. 110. 111. 112. 113. 114. 115. 116. 117. In my childhood I was physically mishandled or abused (other than spanking) by my mother ( circle appropriate answer). Very Often Often Occasionally Seldom Never If YES: In my opinion this happened because: In my childhood I was physically mishandled or abused (other than spanking) by my father (circle appropriate answer). Very Often Often Occasionally Seldom Never If YES: In my opinion this happened because: If ever injured, the injuries were: (circle as many as apply to you): a Bruises e. Red marks b Cuts f. Broken bones c. Cuts requiring stitches 9. Permanent disfigurement d Welts h. Other As a child I saw my father physically mishandle or abuse my mother: Very Often Often Occasionally Seldom Never If YES: In my opinion this happened because: As an adult, I have been physically mishandled or abused by my husband: Very Often Often Occasionally Seldom Never If YES: In my opinion this happened because: 104 12 118. If YES, when l was physically mishandled or abused by my husband: a. I did nothing to deserve it b. I seldom did anything to deserve it c. I occasionally did something to deserve it d. I often did something to deserve it e. I always did something to deserve it. 119. If YES, I received medical attention for the injuries sustained: a. Yes b. No 120. If ever injured, the injuries were: (circle as many as apply to you). a. Bruises e. Red marks b. Cuts f. Broken bones c. Cuts requiring stitches 9. Permanent disfigurement d. Welts h. Other ANSWER IF YOU WERE PHYSICALLY MISHANDLED OR ABUSED: 121. How often was drinking involved when you were mishandled or abused by: Very Qfien Often Occasionally Seldom Never a. Mother A B C D E b . Father A B C D E Husband A B C D E 122. How often were you drinking when you were physically mishandled or abused by: Very Often Often Occasionally Seldom Never a. Mo ther A B C D E b . Father A B C D E c. Husband A B C D E 105 13 The next set of questions ask you to describe yourself. Each item has a scale marked with the letters A, B, C, D, and E with (A) indicating ”Very Characteristic of Me" and (E) indicating "Not At All Character- istic of Me, ” and the other letters are points in between. For each item, circle the letter which best describes how character- istic the item is of you. Refer to this scale when answering this set of questions: Very Much Not At All Characteristic Characteristic of Me F_ai5ly Slightly Not Very of Me A B C D E A B C D E 123. I am unable to relax A B C D E 124. When people are nice to me, it is generally because I have done something to make them that way. A B C D E 125. I don't like weekends or vacations. A B C D E 126. I make friends easily. A B C D E 127. l have no trouble keeping jobs. A B C D E 128. I get the respect I deserve in this world. A B C D E 129. l have financial problems. A B C D E 130. I experience excessive sweating. A B C D E 131. I have sexual problems. A B C D E 132. l have a good memory. A B C D E 133. I am shy with people. A B C D E 134. When someone gets mad at me, I can usually do something to make him my friend again. A B D E 135. I am overambitious. A B D E 136. I find it hard to know whether or not a person really likes me. A B C D E 137. I have inferiority feelings. A B C D E 138. I am lonely A B C D E 139. My home conditions are bad. A B C D E 140. It is easy for me to have a good time. A B C D E 141. When people don't like me it is because I don't know how to get along with them. A B C D E 142. It is easy for me to concentrate. A B C D E 143. When I get into an argument, it is sometimes my fault. 106 14 The next set of questions also ask you to describe yourself. Each item has a scale from A to E and you are to choose the letter which comes closest to describing you. EXAMPLE: Beautiful A B C D E Ugly If you feel beautiful, you would choose A If you feel sanewhat beautiful, choose B If you feel neither beautiful nor ugly, choose C If you feel somewhat ugly, choose D If you feel very ugly, then choose E 144. Worthwhile A B C D E Worthless 145. Useful A B C D E Useless 146. A "somebody" A B C D E A "nobody" 147. "Life is full" A B C D E "Life is empty" 148. Adequate A B C D E Inadequate 149. Smart A B C D E Stupid 1 50. Competent A B C D E Incompetent 151. "Up on things" A B C D E Naive 152. "Can do everything A B C D E "Can't do any- right" thing right" 153. Not at all guilty A B C D E Very guilty 154. Morally right A B C D E Morally wrong 155. Relaxed A B C D E Anxious 156. Calm A B C D E Agitated 157. B rave A B C D E Cowardly 158. Assertive A B C D E Unassertive 159. Passive A B C D E Aggressive 160. Beautiful A B C D E Ugly 161 . Attractive A B C D E Unattractive 162. Appealing A B C D E Repulsive 107 ' 15 SEXUALITY INFORMATION 163. My present sex life is satisfactory with my husband. All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 164. While I am having sex with my husband, I am active. All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 165. It is important to: me in having good sexual experiences to know that my husband cares about me. All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 166. My best sexual experiences with my husband occur when I am feeling playful. All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 167. (ANSWER IF APPROPRIATE) As a child I was sexually molested by ( circle appropriate answers) a. Babysitter (male) 9. Neighbor girl b. Babysitter (female) h. Neighbor (adult male) c. Brother i. Neighbor (adult female) d. Sister J. Stranger e. Father k. Other f. Neighbor boy 168. If you were sexually molested as a child, please describe the circumstances: 168. l have been raped or sexually molested as an adult: 0. Yes b. No 170. If YES, please describe the circumstances. 108 MARITAL HISTORY (where appropriate) 171. 172. 173. I74. I75. I76. I77. 178. 16 How long did you know your marriage partner before engagement? ( circle closest answer) 109 More Than Days Weeks Months 1 Year 1 Year A B C D E How long have you been married: I have pleasant thoughts about my husband: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E I have horrible thoughts about my husband: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E I feel friendly toward my husband: All of Most of Some of the Time the Time the Time Seldom Notat All A B C D E I feel hostile toward my husband: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E I feel full of love toward my husband: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E I feel full of hate toward my husband: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 17 DECISION MAKING 179. 180. 181. 182. 183. My husband and I agree on decision we make: % of time) 100 75 50 25 0 In my marriage I make decisions about how Money Will be Seent: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E In my marriage I make decisions about whom WE will spend time with: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E In my marriage I make decisions about how we will spend OUR free time TOGETHER: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E In my marriage I make decisions about how the children are to be punished: All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E 110 MARITAL RELATIONSHIP Most persons have disagreements in their relationships. cate below the approximate extent of agreement or disagreement between you and your partner for each item on the following list: 184. 185. 186. 187. 188. 189. 190. 191. Handling family finances : Always Usually Agree Half Agree Agree of the Time A B C Matters of recreation: Always Usually Agree Half Agree Agree of the Time A B C Religious matters: Always Usually Agree Half Agree Agree of the Time A B C Demonstrations of affection Always Usually Agree Half Agree Agree of the Time A B C Friends Always Usually Agree Half Agree Agree of the Time A B C Sex relations Always Usually Agree Half Agree Agree of the Time A B C Seldom A gree D Seldom A gree D Seldom A gree D Seldom A gree Seldom A gree D Seldom A gree D Conventionality (correct or proper behavior) Always Usually Agree Half Agree Agree of the Time A B C Philosophy of life: Always Usually Agree Half Agree Agree of the Time A B C 111 Seldom A gree D Seldom A gree D Please indi- Never A gree Never A gree Never A gree Never A gree Never A gree Never A gree Never A gree Never A gree E Marital Relationships (cont.) 192. Ways of dealing with parents or in-laws: Always Usually A gree Half Seldom A gree A gree of the Time A gree A B C D 193. Aims, goals, and things believed important: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 194. Amount of time spent together: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 195. Making major decisions: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 196. Household tasks: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 197. Leisure time interests and activities: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 198. Career decisions: Always Usually Agree Half Seldom Agree Agree of the Time Agree A B C D 112 Never A gree Never A gree Never A gree Never A gree Never A gree Never A gree Never A gree E 20 Marital Relationsmps (cont.) 199. 200. 201 . 202. 203. 204. 205. 206. Raising the children: Always Usually A gree Half Seldom Never A gree A gree of the Time A gree A gree A B C D E How often do you discuss or have you considered divorce, separation, or terminating your relationship? All of Most of the Time the Time Occasionally Rarely Never A B C D E How often do you or your mate leave the house after a fight? All of Most of the Time the Time Occasionally Rarely Never A B C D E In general, how often do you think that things between you and your partner are going well? All of Most of the Time the Time Occasionally Rarely Never A B C D E Do you confide in your mate? All of Most of the Time the Time Occasionally Rarely Never A B C D E Do you ever regret that you married? (or lived together)? All of Most of the Time the Time Occasionally Rarely Never A B C D E How often do you and your partner quarrel? All of Most of the Time the Time Occasionally Rarely Never A B C D E How often do you and your mate "get on each other's nerves?" All of Most of the Time the Time Occasionally Rarely Never A B C D E 113 21 Marital Relationships (cont.) 207. Do you kiss your mate? Almost Every Day Every Day Occasionally Rarely Never A B C D E 208. Do you and your mate engage in outside interests together? Almost Every Day Every Day Occasionally Rarely Never A B C D E HOW OFTEN WOULD YOU SAY THE FOLLOWING EVENTS OCCUR BETWEEN YOU AND YOUR MA TE? 209. Have a stimulating exchange of ideas: Never Monthly Weekly Daily More Often A B C D E 210. Laugh together: Never Monthly Weekly Daily More Often A B C D E 211. Calmly discuss something: Never Monthly Weekly Daily More Often A B C D E 212. Work together on a project: Never Monthly Weekly Daily More Often A B C D E 114 22 Marital Relationships (cont.) THESE ARE SOME OF THE THINGS ABOUT WHICH COUPLES SOMETIMES AGREE AND SOMETIMES DISA CREE. Indicate how often either item below caused differences of opinion or were problems in your relationship during the past few weeks. 213. Being too tired for sex: Very Frequently Frequently Occasionally Rarely Never 8 D E 214. Not showing love. Very Frequently Frequently Occasionally Rarely Never A B D E 214. The dots on the following line represent different degrees of happiness in your relationship. The middle point, "happy," represents the degree of happiness of most relationships. Please circle the dot which best describes the degree of happi- ness, all things considered, of your relationship. Very Mostly Happy Most Mostly Very Unhappy Unhappy of the Time Happy Happy 215. Which of the following statements best describes how you feel about the future of your relationship? I want desperately for my relationship to succeed, and would go to almost any length to see that it does. I want very much for my relationship to succeed, and will do all I can to see that it does. I want very much for my relationship to succeed, and will do my fair share to see that it does. It would be nice if my relationship succeeded, but I can't do much more than I am doing now to help it succeed. It would be nice if it succeeded, but I refuse to do any more than I am doing now to keep the relationship going . My relationship can never succeed, and there is no more I can do to keep the relationship going. 115 23 TIME AT HOME 217. 218. 219. 220. 221. When it comes to how much time I spend at home, we agree All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E When it comes to how much time my spouse spends at home, we agree All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E When it comes to spending time at home, I spend More Than Right Less Than T00 Too Much I Should Amount I Should Little A B C D E I feel that my husband restricts the amount of time that I can be with my friends All of Most of Some of the Time the Time the Time Seldom Not at All A B C D E When it comes to spending time at home my husband spends More Than Right Less Than T00 Too Much He Should Amount He Should Little A B C D E 116 24 PLEASE INDICATE HOW WELL EACH OF THE FOLLOWING CHARACTERISTICS DESCRIBE YOU AS YOU NOW SEE YOURSELF. Use the 7 point scale noted. ( 1) would mean that ”This is never, or almost never true”; ( 7) means ”this is always or almost always true" 1 2 3 4 5 6 7 222. 1 2 3 4 5 6 7 223. 1 2 3 4 5 6 7 224. 1 2 3 4 5 6 7 225. 1 2 3 4 5 6 7 226. 1 2 3 4 5 6 7 227. 1 2 3 4 5 6 7 228. 1 2 3 4 5 6 7 229. 1 2 3 4 5 6 7 230. 1 2 3 4 5 6 7 231. 1 2 3 4 5 6 7 232. 1 2 3 4 5 6 7 233. 1 2 3 4 5 6 7 234. 1 2 3 4 5 6 7 235. 1 2 3 4 5 6 7 236. 1 2 3 4 5 6 7 237. 1 2 3 4 5 6 7 238. 1 2 3 4 5 6 7 239. 1 2 3 4 5 6 7 240. 1 2 3 4 5 6 7 241. 1 2 3 4 5 6 7 242. 1 2 3 4 5 6 7 243. 1 2 3 4 5 6 7 244. 1 2 3 4 5 6 7 245. 1 2 3 4 5 6 7 246. 1 2 3 4 5 6 7 247. 1 2 3 4 5 6 7 248. Self-reliant Yielding Helpful Defends own beliefs Cheerful Moody Independent Shy Conscientious Athletic Affectionate Theatrical Assertive Flatterable Happy Strong Personality Loyal Unpredictable Forceful Feminine Reliable Analytic Sympathetic Jealous Has leadership abilities Sensitive to the needs of others Truthful 117 dddd‘d‘d‘ddddddd‘d‘dddddd‘ddddd‘d NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww R-Fk-B-BG-E'uB-9-E-F-flcE-E-ERJ:#RS-P-E-PRR-fl-R-Pfi-fl'k-fl'fi mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm mammmmammmmmommmmmmmmmmmmmmmmmosaic: NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275. 276. 277. 278. 279. 280. 28I. 25 Willing to take risks Understanding Secretive Makes decisions easily Compassionate Sincere Self-sufficient Eager to soothe hurt feelings Conceited Dominant Soft spoken Likeable Masculine Warm Solemn Willing to take a stand Tender Friendly Aggressive IGLflfitfle Inefficient Acts as a leader Clfikflfike Adaptable lndividualistic Does not use harsh language Unsystematic Competitive Loves children Tactful Ambitious Gentle Conventional 118 282. As I took the questionnaire, I wanted to tell you: 119 26 APPENDIX B DISCUSSION OF TRENDS IN THE DATA FOR SECTION 2: LIFE HISTORIES 120 APPENDIX B DISCUSSION OF TRENDS IN THE DATA FOR SECTION 2: LIFE HISTORIES General Hypothesis No difference will be found between battered women and non- battered women with respect to life histories. Findings Although differences at all levels of family history cannot be verified by the data as evidenced by the non-significance of the MANOVA it was decided to look for trends in the subcategories of life history. Whereas differences between mother- and father on their own sex-typed behaviors as well as Father Protectiveness are insignificant, differences do seem consistent on mother protectiveness. Mothers of non-battered women seem to be significantly more harmonious, supportive and demo— cratic. Within life histories at an alpha level of .05, a trend appeared in which Female Family Harmony and Mother Supportiveness were found to be significant (p = .02 with all scales included, and p = .04 with the Father Masculinity and Father Femininity scales deleted) as was Mother Democracy (p = .01 with all scales included and p = .04 with the. two scales deleted). The means were higher on these scales for non-battered women. All of the variables, their means and F and p values can be found in Table 8.1. 121 122 Discussion of Trends It appears that battered women in this study had mothers who provided significantly less family harmony and were less sup- portive of their daughters than the mothers of non-battered women. The mothers of battered women, as compared to the mothers of the non-battered women, were found to be significantly less democratic and provided less family harmony and support for their daughters. No scales were incorporated in this study to measure father family harmony and support. However, fathers of battered women were found to be relatively more democratic compared to the mothers. This finding suggests that the learned helplessness of battered women had its beginnings in childhood with the mother being the significant parent. The childhood experience of an undemocratic, and possibly abusive mother, combined with sex role socialization, provides a basis for feelings of powerlessness and helplessness. When this is combined with repeated beatings by the adult child's husband, she is sapped of any motivation. She begins to feel again, as she did in childhood, that any voluntary behaviors on her part will not change the situa- tion. Women have been socialized into roles which encourage dependency on men and teaches the women to be nurturing, compliant and passive (Walker, 1981, p. 82). This socialization pattern, when combined with the tendencies of the mothers to be less democratic and to provide for less family harmony and support, may increase the 123 TE. 00 _0>0_ 050.0 50 00 505020500 0. 00500....59m. .00. 0.0 _0>0_ 050.0 50 50 5500559m a; N5. 55.. 5.5 55.N t. .5.. 2.5 55.N 208850 .850“. 55. .55 55.5 55.N .5. 55.5 55.5 55.N $2088.50 .5522 55. 55.5 55.5 55.N N5. 5N.5 N55 55.N immacgzcoaaam .5522 .050 55505.50: 2.80“. 0.060”. 5.. 55.N 5N.5 55.5 5.. 55.N N55 55.5 58858.25 2:50“. 55. N. 5.5 55.5 55. N5. .5.N 55.5 5820555050 250 xom 55. NN.5 5N.5 .5.5 25.530502 .850“. 55. 55. 55.5 55.N 2:55.555 .550“. 55. 5555. 5.5 5.5 55. N55. NN.5 5N.5 3.5.3522 .5522 5.. 55.N N..5 55.5 .N. .5.. 5.... 55 .5 3.5.525“. .0522 NN. 5.. 55.. N... .N. 5.. 55.. 5... $522 00535. .550”. 55.. NN.. 55.. 55.. 5.. 55.. 55.. 55.. 8:205 082230 55. 55.5 55.5 55.N 55. 55.5 55.5 55.N 225.85 .8522 NN. 55.. 5N.5 55.N 5.. 55.N 5N.5 55.N 2:258... .650... .5. 55. .5.5 5N.5 55. 555. .5.5 55.5 280.. .8520 .5.uz. Euz. .5Nuz. .572. 0.02._m> a u. 5.02 2.0 a n. 2.02 5.0 500.2 500.2 5000.00 00.00m 5053.05. 00.00m 00.5050... 00.... HN 50.5.00ml. 0050...0> “.0 m_m>_05< 050.502.531.255 wumdfi. 124 dependency on fathers and men and thereby reinforce the learned helplessness syndrome of battered women. The borderline personality and object-relations theory pro- vides another framework for the battering situation when considering the finding that mothers of battered women were less democratic, less supportive, and less likely to provide for family harmony than the mothers of non-battered women. Object relations involves two internal representations, one of the self and one of the significant other, the mother, and an emotional bond between them. During the first two stages of object relations the images of the self and the other are perceived as one. The third stage is marked by the child differentiating self from other but having two separate sets of object relations, one good and one bad. These two object relations are (1) a good, compliant self image who is in a positive, loving relation to a warm, giving, good mother image, and (2) a bad self image in a hateful tie to a depriving or rejecting, bad mother image (Gillman, 1980, p. 347). In the fourth stage, good and bad images come together to form an integrated whole. Like the borderline personality the battered woman never reaches stage four because of the lack of support from her mother, but rather remains split. To quote Irene Gillman (1980, p. 348): The battered woman has two separate and quite distinct dyadic representations: her lovable self in a warm, friendly relation to a good, providing husband-mother, and her helpless, worthless self who is in a hateful, destructive relationship to a persecuting, damaging husband-mother. 125 The battered woman, similar to the borderline personality, may employ the defense mechanisms of splitting and projective identi- fication, introjective identification, and denial. Sometimes the bad object is projected out and the woman feels like the good one with the other person, the husband, being perceived as completely bad. At other times the bad object becomes introjected and may explain why the battered woman, both as a child and as an adult, felt deserving of the punishment she received. This may also help to explain the woman's role in eliciting the beating from her husband since during projective identification the battered woman unconsciously attempts to get the other person to become one with the feelings and behaviors the woman temporarily experiences (Gillman, 1980, p. 348). They are either both good or both bad. Splitting and denial may combine in phase three of the cycle of violence and may offer an explanation for why the woman stays in the relationship. After the beating, love reemerges without contamination from the bad self object representation, a result of splitting and denial. This process was learned as a child in order for the child to survive the feelings of powerlessness stemming from the relationship with her mother. HICH GAN STATE