Você está na página 1de 5

Journal of Consulting and Clinical Psychology 1986, Vol. 54, No.


Copyright 1986 by the American Psychological Association, Inc. 0022-006X/86/J00.75

Long-Term Effects of Sexual Victimization in Childhood: An Attributional Approach

Erica R. Gold
University of Manitoba

This study explored the relation between childhood sexual victimization experience and adult functioning. One hundred and three adult women who were victimized as children or adolescents served as subjects, and 88 women who were not victimized served as controls. Members of both groups completed a questionnaire about their present social, psychological, and sexual functioning. Sexually victimized women also provided information about their victimization experiences. Victimized women differed significantly from non victims on measures of (a) childhood family and social experiences; (b) adult attributional style; and (c) level of depression, psychological distress, self-esteem, and sexual problems. Results of the study suggested that the sexually victimized women's adult functioning was related most strongly to their attributional style for bad events. Perception of the victimization experience and quality of social support were also important factors related to adult functioning. Possible implications of these findings for treating women who are experiencing problems related to childhood sexual victimization experiences are discussed.

Research suggests that as many as one of every four girls in North America may be sexually victimized before she reaches adulthood (e.g., Finkelhor, 1979). Recent studies, which have found that up to 44% of females who have received help in clinical settings were sexually victimized as children, suggest that many victims experience psychological problems (e.g., Briere, 1984). Although most studies have found that female victims exhibit no specific pattern of response to childhood sexual abuse (e.g., Meiselman, 1978), certain symptoms are commonly observed. These include depression, low self-esteem, guilt feelings, and interpersonal difficulties characterized by feelings of isolation and difficulty in trusting others (e.g., Tsai & Wagner, 1978). Studies of the long-term effects of childhood sexual abuse tend to focus on selective subgroups (i.e., university students or clinical samples) and therefore the findings may not be generalizable to the whole population of victimized women. Similarly, studies do not always use adequate control groups, and many studies provide demographic information but do not use statistical techniques (e.g., Meiselman, 1978). Negative effects of abuse may be assessed from subjective reports or from vaguely denned types of pathology exhibited by the victim (e.g., Finkelhor, 1979; Meiselman, 1978). The major focus of research into the long-term effects of

childhood sexual victimization has been on the relationship between variables associated with the victimization situation and the victim's adult functioning. Factors associated with the individual's response to the abuse have rarely been investigated, although many authors have pointed out that examination of the person by situation interaction is a more accurate conceptualization of behavior than separate observation of either personality or situation (e.g., Bowers, 1973). For the present research, it was hypothesized that variables related to the victim, specifically her attributional style, would provide an explanation for victimization effects. The learned helplessness model of depression (Abramson, Seligman, & Teasdale, 1978) was used to provide a conceptual framework for the study. The model postulates that a person's causal attributions and expectancies mediate response to uncontrollable situations. Attributions are categorized according to three dimensions: internal-external, stable-unstable, and global-specific. If childhood sexual victimization is conceptualized as a helplessness experience (i.e., an uncontrollable event), then learned helplessness theory would suggest that the symptoms observed in many victims may result from internal, stable, global attributions for bad events and from expectations of having no control over the environment. The present study was designed to explore the relation between childhood sexual victimization experience and adult psychological and sexual functioning. It was hypothesized that the

This article is based on the author's doctoral dissertation submitted to the University of Manitoba. Research was supported in part by Manitoba Mental Health Research Foundation Grant 387-1665-29 and in part by Medical Research Council of Canada Grant MA5028. The author gratefully acknowledges the assistance of Larry Beutler, Lillian Esses, and Warren Nielson who provided many helpful comments on earlier drafts of this article. Correspondence concerning this article should be addressed to Erica Gold, King's College, 266 Epworth Avenue, London, Ontario, Canada, N6A 2M3.

victims' attributional styles would explain the apparent relation between aspects of the victimization situation and adult functioning. Specifically, an internal, stable, global attributional style for bad events was expected to correlate with poor functioning in victims. Social support was also examined as a potentially important factor mediating the effects of childhood sexual victimization (Silver & Wortman, 1980). High levels of social support were expected to correlate with good functioning, and low levels of social support were expected to correlate with poor


ERICA R. GOLD interviewed about their victimization experience. The purpose of the study was described in greater detail after the tasks were completed, and participants then had an opportunity to ask questions and discuss their responses to the study. Feedback was provided to participants, referring agencies, and clinicians after completion of the study.

functioning (Sarason, Levine, Basham, & Sarason, 1983). Differences between personalities, social support levels, and adult functioning of victimized and nonvictimized women were compared. It was hypothesized that victims, as a group, would not cope as effectively as nonvictims. Victims were expected to have different attributional styles than nonvictims, and these styles were expected to be characterized by more frequent internal, stable, global attributions for bad events.

Demographic information about the participant's family background, personal history, and victimization experience (if any) was collected using a modified version of the questionnaire developed by Finkelhor (1979). Items were changed, added, or deleted to make the questionnaire applicable to a broader sample of women than the university sample used in Finkelhor's research. The following measures were included in the experimental questionnaire to assess the subject's present level of functioning: 1. The Beck Depression Inventory (BDI; Beck, 1978), a clinically derived 21 -item self-report scale, was used to assess depression level. 2. The Hopkins Symptom Checklist (HSCL; Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 1974) provided a self-report of the number and intensity of symptoms experienced. The HSCL measures five dimensions of symptoms (somatization, obsessive-compulsive, interpersonal sensitivity, depression, and anxiety) through ratings on a 4-point scale. 3. The short form of the Texas Social Behavior Inventory (TSBI; Helmreich & Stapp, 1974) was used to assess self-esteem or social competence. This 16-item inventory presents statements that the subject rates on a 5-point scale from not at all characteristic of me (0) to very characteristic of me (4). 4. Some items selected from the Index of Sexual Satisfaction (Hudson, Harrison, & Crosscup, 1981) and some items from the Sexual History Form (Schover, Friedman, Weiler, Heiman, & LoPiccolo, 1982) were used to assess sexual functioning. The following measures were used to assess attributional styles: 1. Attributional style was measured using the Attributional Style Questionnaire (ASQ) developed by Peterson et al. (1982). The scale asks "subjects to generate a cause themselves for each of a number (12) of events and rate the cause along seven point scales corresponding to the internality. stability and globality dimension" (Peterson et al., 1982, p. 289). 2. Self-blame was assessed using the scenarios developed by JanoffBulman (1979). Respondents are asked to imagine themselves as the target person in each of four scenarios and to respond to five questions about the scene. Questions assess locus of blame (self, other, world, and chance), characterological and behavioral self-blame, and deservingness. 3. Locus of control expectancy was measured using the Internal-External Locus of Control scale (Rotter, 1966). This scale is a 29-item forced-choice test (including 6 filler items). Social support was measured using two dimensions of the Social Relationship Scale (SRS; McFarlane, Neale, Norman, Roy, & Streiner, 1981). The respondent was presented with two of five categories of life stress (home and family, and personal and social). She was asked to list the people with whom she had discussed these issues, to rate the helpfulness of the discussions with each person on a 7-point scale, and to note whether the relationship was reciprocal. She then listed the key people to whom she would turn in times of crisis and noted whether these people would reciprocate. The Marlowe-Crowne Social Desirability Scale (Crowne & Marlowe, 1960) was used to assess the tendency of subjects to seek social approval by responding in a culturally appropriate manner.

For the purpose of the study, sexual victimization was defined as (a) sexual contact (i.e., physical touch) between a child (12 years old and under) and a postpubertal person at least 5 years older than the child; (b) sexual contact between an adolescent (13 to 16 years old) and an adult at least 10 years older; or (c) sexual contact between any child or adolescent under 16 years old and a person or persons who used physical force.

One hundred and three women who had been sexually victimized as children participated in the study. The women ranged in age from 18 to 56, with a mean age of 30.4 (SD = 8.6). The control group consisted of 88 women who had not been sexually victimized as children. These women ranged in age from 18 to 37, with a mean age of 29.8 (SD = 8.2). The majority of the participants (83.5% of the victimized women and 79.5% of the nonvictimized women) were recruited through advertisements placed in local and campus newspapers and through notices posted in the community. Other participants (8.7% of the victims and 3.4% of the nonvictims) were recruited through clinicians and social agencies, and the remaining participants (7.8% of the victims and 17% of the nonvictims) were students in introductory psychology classes who were given research credit for their participation. In each group, approximately 43% of the women were single, 30% were married, 25% were separated or divorced, and fewer than 2% were widowed. Approximately 25% of the women in each group were not employed outside the home. Between 22% and 24% of the women earned less than $ 10,000 (Canadian dollars) per year, about 25% earned between $10,000 and $20,000, and between 15% and 20% earned between $20,000 and $40,000. About 20% of the victims and 5% of the nonvictims had attended or completed high school, 11-13% of each group had completed high school plus some other training, and approximately 55% of each group had attended some college or had completed college. Fifteen percent of the victims and 28% of the nonvictims had a professional degree, graduate training, or a graduate degree. To determine whether the two groups differed significantly in age, education level, individual income and husband's income, and social desirability set, t tests were conducted on these variables. A significant difference was found between the groups in education level, with nonvictims found to be more educated than victims, ((188) = -3.02, p < .003. No significant differences were evident for any of the other variables.

Prospective participants contacted the researcher or research assistant by telephone and an appointment was scheduled. A brief description of the study was provided during this initial contact, and participants were introduced to the research in greater detail when they met individually with the experimenter. They were advised that they could terminate their participation at any time and were assured of the confidentiality of their responses. Informed consent was obtained. All participants completed an experimental questionnaire, and sexually victimized women, identified by their responses to the questionnaire, were



Results The Victimization Experiences

Table 1 Significant Univariate F Ratios Variables

F(l, 164)

The 103 women who were sexually victimized as children or adolescents reported a total of 191 victimization experiences with different offenders that met the criteria for inclusion in this study. The number of different offenders involved with each victim ranged from 1 to 5 (M = 1.84, SD = 1.02). The age of the victim at the time of the abuse ranged from 2 to 16. The mean age was 9.74 (SD = 3.79). In 5 1 .3% of the incidents the offender was an acquaintance of either the victim or her parents, in 36.3% of the incidents he was a family member, and in 12.2% of the incidents he was a stranger. When the offender was a family member, 31.9% of the incidents occurred with the victim's father, 21.7% with an uncle, 1 7.4% with a brother, and 1 3% with a grandfather.1

Between-Group Analysis
A multivariate analysis of covariance, with education level as the covariate, was conducted to compare the victims with nonvictims on 42 measures of family violence, social support, attributional style, and adult functioning. Data from 167 subjects (9 1 victims and 76 nonvictims for whom no relevant information was missing) were included in the analysis. Comparison of these two groups of women using t tests revealed no significant differences in measures of age, individual and husband's income, and social desirability set. Education level, the covariate, was different for the two groups, t(\65) = -3.30, p < .001. There was a significant main effect for group, f\l, 123) = 2.28, Univariate F tests indicated significant differences between the groups for many of the variables (see Table 1 ). As compared with the nonvictims, victimized women reported fewer friends at age 12, less closeness with their parents when they last lived with them, more family violence, and more physical punishment at age 12 from their parents. At the time of the study, victims were more likely than nonvictims to attribute bad events to global factors and to internal, stable, global factors; to blame their character and behavior for bad events; and to attribute good events to external factors. Victims were less likely to blame other people for bad events. Victims, as compared to nonvictims, also reported more negative sexual symptoms, fewer positive responses to sexual invitations, less satisfaction with present sexual relationships, more depression, a greater number of and more intense psychological symptoms, and lower self-esteem. Victims were more likely than nonvictims to report that they had received psychotherapy or had been hospitalized for a psychiatric problem.

Psychological help received Number of friends (age 12) Closeness to mother Closeness to father Overall family violence Spanking by mother Spanking by father Globalbad events (ASQ) Characterological blame Internal, stable, globalbad events (ASQ) Internalgood events (ASQ) Other blame Behavioral blame Depression (BDI) Intensity of symptoms (HSCL) Psychological symptoms (HSCL) Negative sexual symptoms Sexual responsiveness Self-esteem (TSBI) Sexual satisfaction

18.96' 16.06' 14.70' 7.97' 7.90' 7.27' 6.56* 11.82"* 10.23** 6.80** 5.79* 4.19* 3.98* 17.33"** 11.63*** 11.44*" 11.16*** 7.70** 5.97* 5.11*

Note. n = 9\ victims; n = 76 nonvictims. ASQ = Attributional Style Questionnaire; BDI = Beck Depression Inventory; HSCL = Hopkins Symptom Checklist; TSBI = Texas Social Behavior Inventory. *p<.05."p<.01***p<.001.***p<.OOOI.

behavioral self-blame measures) was related to the adult functioning variable, which had a high positive loading on psychological distress (BDI and HSCL scores combined) and a high negative loading on TSBI. The second relationship, K36) = .60, p < .0001, indicated that high positive loading on negativity of victimization experience and on ASQ total score for good events was related to an adult functioning variable that had high positive loadings on both psychological distress and TSBI. The third relationship, r(22) = .55, p < .01, indicated that high negative loading on reciprocity and helpfulness of social supports, closeness to mother, and attributional style for bad events was related to an adult functioning variable defined by a high positive loading on sexual problems. Relationship of attributional style and social support with adult functioning: Nonvictims. One significant relationship emerged from the canonical correlation of these variables, r(32) = .57, p < .005. The attribution and social support variable was defined by high positive loadings on attributional style for bad events and psychological help received and by high negative loadings on ASQ total score for good events and reciprocity and helpfulness of social supports. The variable was related to the adult functioning variable that had a high positive loading on psychological distress (see Table 3).

Within-Group Analysis
Relationship of victimization experience, social support, and attributional style with adult functioning: Victims. Three significant relationships emerged from the canonical correlation of these variables (see Table 2). The first relationship, n[52) = . 82, p < .000 1 , indicated that a high positive loading on attributional style for bad events (a score composed of the ASQ total score for bad events combined with the characterological and

The present study sought to improve upon past research into the long-term effects of sexual victimization of children. The

1 Further descriptive data about the victimization experiences are available from the author.

Table 2 Canonical Correlations of Combined Variables and Adult Functioning: Victims First canonical variable Second canonical variable

ER]CA R. GOLD sexual relationships. Girls who were close to mother at age 12 may have had mothers who were abused themselves and therefore had difficulty with their sexual functioning and mothering Third canonical variable skills. These mothers may have had difficulty both allowing their daughters to separate from them and providing their daughters with an appropriate role model for positive social and sexual functioning. The implications of the relationship between the victimized girl and her mother remain unclear and
.294 .119 -.062 .166 -.225 -.506 -.015 -.018 .082 -.480 .119 -.580 -.373


Victimization, attribution^ style, social support Negativity Age difference Force Relationship ASQ totalgood Attribution bad Locus of control Response to victim Psychotherapy Reciprocity/helpfulness Closeness to father Closeness to mother Number of supports
-.025 -.123 -.057 -.092 -.136 .672 .206 -.073 .074 -.204 .095 .001 -.248

.625 .065 -.115 .170 .447 .081 -.150 .259 .333 -.193 -.381 .092 .251

warrant further investigation. At this time one cannot determine whether the victimized woman's attributional style developed prior to the victimization or in response to it, or whether there is an interaction between attributional style and sexual victimization experience. The finding that victimization variables and attributional style variables were not significantly correlated suggests that attributional style may be independent of the victimization experience. However, the importance of the victimization experience is indicated by the different relationships found between attributional style and coping effectiveness in victims and nonvictims. The adult functioning of nonvictims was related to attributional style and type and quality of social support, whereas the

Adult Functioning Self-esteem (TSBI) Sexual problems Distress (BDI + HSCL) Antisocial behavior
-.497 .007 .593 -.118 1.147 .105 1.109 .019

.141 .971 -.273 -.236

functioning of victims was strongly correlated with attributional style alone. Clearly, further research is in order to elucidate this relationship. Although the sample was large and an attempt was made to recruit a cross-section of women to participate in this study, the generalizability of the findings to other sexually victimized women would be limited if the volunteers were not representative of the populations of victimized and nonvictimized women. Causal statements cannot be made about the relationships among variables in this study because (a) in retrospective studies such as this one it is not possible to control for either the effects of events which intervened between victimization and present assessment or for pre-existing psychological distur-

Note. ASQ = Attributional Style Questionnaire; BDI = Beck Depression Inventory; HSCL = Hopkins Symptom Checklist; TSBI = Texas Social Behavior Inventory.

coping effectiveness, social support systems, and attributional styles of a large community sample of women who were sexually victimized as children were compared with those of a similar control group of women who were not victimized. As hypothesized, victimized and nonvictimized women showed significant differences on many measures of attributional style, social support, and coping effectiveness. For the first time, aspects of the woman's victimization experience, attributional style, and social support network were related to her adult functioning. The most significant finding to emerge from this study was the close relationship between the victim's attributional style and her adult functioning. The results indicated that, as predicted, women who were sexually victimized in childhood and who reported psychological distress and low self-esteem were likely to display an attributional style marked by internal, stable, global attributions for bad events. The results raise the possibility that the reported coping difficulties of some victims, especially distress and low self-esteem, may be due to the attributions that they make for bad events. The victimized woman's present perception of the abusive experience and her perception of her mother's response to it were the only aspects of the victimization experience that were significantly related to her adult functioning. Two aspects of social support were found to be related to victims' adult functioning, or specifically, to their sexual functioning. Women who reported good quality adult social relationships (helpful and reciprocal) also tended to report satisfactory sexual relationships. The quality of the support reported by these women may reflect their ability to have close relationships with others, including

Table 3 Canonical Correlations of Combined Variables and Adult Functioning: Nonvictims First canonical variable


Attributional style, social support ASQ totalgood Attributionbad Locus of control Psychotherapy Reciprocity/helpfulness Closeness to father Closeness to mother Number of supports Adult functioning Self-esteem (TSBI) Sexual problems Distress (BDI + HSCL) Antisocial behavior
-.163 .281 .840 .041 -.301 .523 .012 .374 -.342 .032 .176 .230

Note. ASQ = Attributional Style Questionnaire; BDI = Beck Depression Inventory; HSCL = Hopkins Symptom Checklist; TSBI = Texas Social Behavior Inventory.

EFFECTS OF CHILDHOOD SEXUAL VICTIMIZATION bances; (b) subjective data were used that may be purposefully or unknowingly distorted due to memory impairment or discomfort with the subject matter; and (c) correlational procedures were used. The differences observed in attributional styles and reported functioning of victims and nonvictims may be a function of the differences between groups in the number of women who reported receiving psychotherapy. These differences might not be observed if victims were compared with nonvictims who have received therapy. The interview data suggested that many sexually victimized women felt that they had difficulty developing trusting relationships. Therefore, it may be particularly important for a therapist working with such a client to provide a supportive environment and to listen to the client in an empathic, nonjudgmental way. Given that in victims a self-blaming pattern was related to reports of depression, psychological distress, and low selfesteem, cognitive therapy techniques could be used to explore attributional style and to identify and modify self-defeating thoughts (see Beck, Rush, Shaw, & Emery, 1979). Group therapy may be useful for reducing identified feelings of isolation and/or guilt, for increasing self-esteem, and for increasing the ability to trust other people (e.g., Tsai & Wagner, 1978). The sexual problems of many victimized women appeared to be related to trust and intimacy problems, and these may improve through the normal course of therapy. It has been suggested that the therapist working with such women focus on the cognitive factors that affect the victim's perceptions of sexual stimuli rather than on physiological responses per se (Becker, Skinner, Abel, & Treacy, 1982). It may also be important for the therapist to explore the woman's relationship with her parents, especially her mother.


Abramson, L., Seligman, M., & Teasdale, J. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74. Beck, A. (1978). Beck Inventory. Philadelphia, PA: Center for Cognitive Therapy. Beck, A., Rush, A., Shaw, B., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. Becker, J., Skinner, L., Abel, G., & Treacy, E. (1982). Incidence and types of sexual dysfunctions in rape and incest victims. Journal of Sex and Marital Therapy, 8, 65-74.

Bowers, K. (1973). Situationalism in psychology: An analysis and a critique. Psychological Review, 80, 305-336. Briere, J. (1984, April). The effects of childhood sexual abuse on later psychological functioning: Defining a post-sexual abuse syndrome. Paper presented at the Third National Conference on Sexual Victimization of Children, Children's Hospital National Medical Center, Washington, DC. Crowne, D., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. Derogatis, L., Lipman, R., Rickels, K., Uhlenhuth, E., & Covi, L. (1974). The Hopkins Symptom Checklist (HSCL): A self-report symptom inventory. Behavioral Science, 19, 1-15. Finkelhor, D. (1979). Sexually victimized children. New York: Free Press. Helmreich, R., & Stapp, J. (1974). Short form of the Texas Social Behavior Inventory (TSBI), an objective measure of self-esteem. Bulletin of the Psychonomic Society, 4, 473-475. Hudson, W., Harrison, D., & Crosscup, D. (1981). A short-form scale to measure sexual discord in dyadic relationships. Journal of Sex Research, 17, 157-174. Janoff-Bulman, R. (1979). Characterological versus behavioral selfblame: Inquiries into depression and rape. Journal of Personality and Social Psychology, 37, 1798-1809. McFarlane, A., Neale, K.., Norman, G., Roy, R., & Streiner, D. (1981). Methodological issues in developing a scale to measure social support. Schizophrenia Bulletin, 7, 90-100. Meiselman, K_ (1978). Incest: A psychological study of causes and effects with treatment recommendations. San Francisco: Jossey-Bass. Peterson, C., Semmel, A., von Baeyer, C., Abramson, L., Metalsky, G., & Seligman, M. (1982). The attributional style questionnaire. Cognitive Therapy and Research, 6, 287-300. Rotter, J. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80, 1-28. Sarason, L, Levine, H., Basham, R., & Sarason, B. (1983). Assessing social support: The social support questionnaire. Journal of Personality and Social Psychology, 44, 127-139. Schover, L., Friedman,!., Weiler, S., Heiman, J., & LoPiccolo, J. (1982). Multiaxial problem-oriented system for sexual dysfunctions: An alternative to DSM-III. Archives of General Psychiatry, 19, 614-619. Silver, R., & Wortman, C. (1980). Coping with undesirable life events. In J. Garber & M. Seligman (Eds.), Human helplessness: Theory and applications (pp. 279-340). New York: Academic Press. Tsai, M., & Wagner, N. (1978). Therapy groups for women sexually molested as children. Archives of Sexual Behavior, 7,417-427. Received December 4, 1984 Revision received September 4, 1985