|
The Measurement of Attribution of Battering: A Review of the LiteratureJ. Michael Norgaard, LPC, NCC Updated February, 2005 The Measurement of Attribution of Battering: A Review of the Literature Family violence is receiving increased attention both in the research community and from the popular media. Violence in the United States occurs more within the family than anywhere else (Strauss, Gelles, & Steinmetz, 1980). In recognizing it's importance, the National Institute of Justice has identified violence against women as a research priority area in women’s health (O’Brien & Murdock, 1993). In the United States as many as three million women are victims of violent crime perpetrated by an intimate partner annually (Petretic-Jackson, Sandberg, & Jackson, 1994). Other less personal methods of data collection, such as questioning respondents about their neighbors’ domestic violence (Pacquin, 1994), suggested that three million may be a gross underestimate of the extent of the problem. Although these figures are disturbing, more alarming is the fact that as many as 99% of all battering incidents go unreported (Sigler, 1989). Of all violence committed against women, almost 30% is committed by intimate partners (Bachman & Saltzman, 1995). Also, 15% of all homicide cases are a result of one family member killing another (Chadwick & Heaton, 1992). In many cases these acts are considered to be justified not only by perpetrators, but also by witnesses (Greenblat, 1985) and perhaps most tragically, in some cases, even by the victim herself (Howard, 1984, a, b). Attribution In 1739, Hume proposed that making causal assumptions about activities occurring in our environment is an automatic response, because it helps add meaning to what is happening around us. The causal assumptions ultimately affect the way in which we feel, react, and behave (Tucker-Ladd, 1996). Schema are attributional shortcuts that are developed in response to an individual's experience. These shortcuts are applied in assigning causal judgments to future situations that are thought by the individual to resemble those of past experience. Once developed, these schema guide attributions (Jones, Kanouse, Kelley, Nisbett, Valins, & Weiner, 1972). The purpose of these attributional shortcuts is to help make sense of the world in a more systematic, expedient manner. The problem that can arise from the use of these preconceived attributions is that they may leave an individual less open to change or consideration of other perspectives. Once this occurs, a risk of becoming locked into outdated belief systems that become self-perpetuating through their repeated application to events in the environment develops. No longer is it acceptable for a brutal, violent act in which serious injury often occurs to be considered anything less than criminal. In more severe cases of attributional error, the women are blamed for the attack, thus being revictimized (Howard, 1984{a, b}, Luginbuhl & Mullin, 1981). These issues are at the heart of studies about attitudes, perceptions, and attributions about violence against women. Societal attitudes about wife battering have an appalling history. In the Middle Ages, for example, wife battering was an expected social norm whose violation was unconscionable. During the early twentieth century, battering was a generally accepted practice that was neither officially endorsed nor condemned. In contemporary United States society the emphasis has gradually been redefined by society as a social problem needing attention and intervention (Frieze, 1979). Although there have traditionally been cultural variations with regard to attitudes about wife battering (Davis, 1992), a majority of societies studied disapproved of the practice to varying extents. Measurement Measuring how people judge, perceive, and attribute blame in battering situations is a critical concept to the growing body of literature about domestic violence. Scores on attitudinal measures can be related to behavioral manifestations of battering (Williams, 1990). Attributions also affect other key concepts such as intervention decisions by police officers (Stewart & Maddren, 1997; Finn & Stalans, 1995; Stalans & Finn, 1995; Lavoie, Jacob, Hardy, & Martin, 1989; Waaland & Keeley, 1985), the likelihood of a conviction of a batterer (Hiller, Foddy, 1993), and sentencing decisions (Lamb & Keon, 1995; Burke, Ames, Etherington, & Pietsch, 1990; Edwards, 1987). Medical practitioners’ decision making processes were related to reporting and intervention and were also related to attributions (Tewksbury & Adkins, 1991), as were treatment decisions by, and attitudes of, social service providers (O’Brien & Murdock, 1993; Davis, 1984; Davis & Carlson, 1981). Attribution of blame may also affect a victim’s decision to leave or remain in an abusive relationship (Arias, 2000).Ultimately, public perceptions and attributions drive political platforms, public policy, and resource allocation (Davis, Hagen, & Early, 1994; Gelles & Loseke, 1993). Despite the importance of how blame is assigned, few attempts have been made to systematically develop and validate empirically based instruments to measure attribution of blame in battering cases. A review of the literature reveals that it is not uncommon to find modified versions of validated instruments reconfigured to suit the needs of a study, or worse, the use of a list of questions developed in seemingly less than systematic fashion with no attempt at validation, norming, or standardization. This paper will examine instrumentation related to attitudes and attribution toward battering. Existing scales, an examination of results, and an examination of factors related to the measurement of these concepts will be covered. Traditionally, three basic approaches have been used to measure attitudes and attribution of violence. Some research has measured attribution using a dichotomy with perpetrator blame on one end and victim blame on the other (Kalmuss, 1979), or having the respondent rate the degree of "blameworthiness" of either the victim or the perpetrator (Kanekar & Vaz, 1988, Luginbuhl & Mullin, 1981). As a more thorough understanding of the dimensions involved in attribution began to develop, theorists learned that the same individual may hold both the victim and the perpetrator independently responsible to some extent. In response, researchers began measuring victim blame and perpetrator blame on separate dichotomies (Bryant and Spencer 2003, Stewart & Maddren, 1997; Waaland, Keeley, 1985; Acock & Ireland, 1983). More contemporary models of attribution of battering make a more concerted effort to capture its multidimensional nature. Typically, these models include not only separate measures of victim and perpetrator blame, but also attempt to capture blame attributed to environmental (societal, media) or situational (relationship, life circumstances) variables (Bryant and Spencer 2003, Lavoie, Jacob, Hardy, & Martin, 1989; Jackson, Dienst, Efird, Mobley, Schroeder, Hout, Montecillo, & LaBine, 1994; Petretic-Jackson, Sandberg, & Jackson, 1994; Saunders, Lynch, Grayson, & Linz, 1987; Howard, 1984{a}, {b}). Dutton (1999) also makes a point that battering may be more appropriately conceptualized as a developing process over time rather than simply a discrete event or series of discrete events. The argument is made that to measure dimensions of battery without appreciating the context in which it occurs may serve to deconceptualize the very construct we seek to understand more thoroughly. Measuring something as complex as human behavior with a unidimensional tool is inadequate because a single dimension fails to adequately consider other variables involved in the attribution process. A multidimensional tool is needed to capture the most comprehensive illustration of the factors affecting attribution in battering cases. The first to suggest a multidimensional approach to measurement of attribution of violence was Brodsky (1976). Howard (1984{a}) suggested that attribution is shaped by cognitive schema developed as a result of experience and socialization. Frequently, researchers neglect to collect demographic variables other than gender of the respondent. Failing to consider exposure to major life impacting experiences, such as level of education, degree of religiosity, or living in poverty, may result in ignoring information paramount to the development of an individual’s perspective of human behavior. Another illustration for the need for a multidimensional measure was made by Petretic-Jackson et al. (1994) when they stated, “because behavioral self-blame is a common reaction, assessing attitudes regarding the relative importance of partner, self, situational, and societal factors in motivating abuse would be valuable in evaluating such self-blame” (p. 267). Another factor that appears to be relevant for report and non-report of violence is race. Brent-Goodley (2004) looked at perceptions of domestic violence among African American women utilizing a series of focus groups. Findings were that African American Women are less likely than other groups of women to report violence perpetrated against them. Possible rationales included fear of victim blame and perhaps incarceration, since African American women are incarcerated for family violence at a higher rate than other groups of women, or that perhaps a perception of harsher treatment of the perpetrator if he were to enter the justice system prevents higher rates of reporting. Brent-Goodley (2004) points out that there are also some cultural influences against reporting such violence that she refers to as Racial Loyalty. African American women are more likely to kill a partner than White females and are twice as likely to be killed by a partner. Brent-Goodley (2004) suggests that this may be due to a hesitancy to seek help and a failure to recognize or acknowledge the level of danger in a violent relationship. Implications of this study are that a cultural understanding of attitudes toward violence are important and that educational campaigns delivered to at risk populations are important before positive change can be expected. Instruments In an earlier study of attribution in wife abuse cases, Kalmuss (1979) used a mail back questionnaire distributed to a sample of more than 370 residents of Detroit over the age of 18. Each respondent read a vignette involving wife abuse with a manipulated past-pattern of abuse, justification, and consequences of the action. The measure of attribution was a split responsibility scale that asked the respondent to assign blame to either the female or the male. The response was indicated on a 9-point Likert scale with “Female totally responsible” and “Male totally responsible” as the anchors, and “Both equally responsible” assigned to the modal value of 5. In the results, 73% of respondents attributed responsibility to the husband, compared with 23.7% sharing blame evenly and 3.3% blaming the wife. Kalmuss (1979) found statistically significant support for her hypothesis that men would assign less responsibility to the husband, but failed to confirm the opposite, that women would assign less responsibility to the wife. In another study, Davis and Carlson (1981) examined the attitudes of 480 service providers toward victims of wife battering. The sample included hospital staff (n=95), police personnel (n=119), family service agency workers (n=123), and domestic violence hotline and shelter staff (n=21) among others. Each respondent was exposed to one of four vignettes of a woman seeking emergency medical treatment for injuries resulting from a beating by her husband. The vignettes varied with regard to socioeconomic status and degree of injury. Respondents completed a brief questionnaire asking them for assignment of responsibility to both the husband and the wife, and degree of sympathy toward each. Subjects also responded on a nine-point Likert scale asking how much they believed personal and behavioral characteristics of the husband and wife contributed to the beating, in addition to stressful events as causal factors. Results showed that while husbands were held more responsible for the abuse, the wives were still assigned some degree of blame by most, with a mean of 5.5 on a 9 point scale. Respondents working in domestic violence shelters held women least responsible, while medical workers assigned them the highest degree of blame, although no breakdown was given regarding the differences, if any, in responses between nurses and physicians. Less educated respondents showed a tendency to hold the husband less responsible (Davis, Carlson, 1981). In a similar study, Davis (1984) surveyed beliefs service providers held toward battered women and abusing men. The sample included subjects representing 24 organizations likely to deal with victims of battering including police officers (n=121), nurses (n=55), physicians (n=41), family court judges (n=20), family service workers (n=69), and hospital social workers (n=37) among others. Each respondent was exposed to one of four emergency room scenarios describing a woman who was injured by her husband. The scenarios varied with regard to socioeconomic status and degree of injury. Respondents answered four open ended questions about factors that would prevent the woman from experiencing future incidents and listed recommendations for each spouse. Factors identified for the wife being at higher risk of future victimization included lack of resources, lack of services, fear, and personality problems ascribed to the wife, followed closely by personality problems ascribed to the husband. Identified barriers for the husband in remaining nonviolent included denial, lack of self- control, lack of knowledge of alternatives, and a habitual abusive behavior pattern. Recommended actions for the wife included counseling and separation or divorce. Recommendations for husbands, however, focused almost exclusively on counseling services. O’Brien and Murdock (1993) also conducted a investigation studying shelter workers’ perceptions of battered women. The sample consisted of 24 female and 1 male staff members including counselors, managers, and one executive director. Materials included four intake forms and a perceptual measure. The forms varied situational factors with regard to stability of cause and locus of cause, with a 2x2 matrix combining external/stable and internal/unstable. The perceptual measure, with a reported internal consistency of .70 across scenarios, consisted of seven items focusing on attitudes about the client. Results indicated that shelter workers’ perceptions of the victim were influenced by the attributional presentation by the victim. Another study (Home, 1994) compared attributions of police officers and social workers in cases of wife abuse. Each of the 235 police officers and 188 social workers were given a vignette describing a domestic violence situation. Eight variations were presented manipulating level of provocation. Each respondent was asked to attribute level of culpability to the perpetrator, the victim, and the socioeconomic situation. Social workers and police officers differed significantly on their attribution. Both samples held the man responsible, but officers were more likely to hold the woman responsible at some level and to blame financial stressors. Police officers were also more likely to minimize the level of seriousness of the situation than either male or female social workers. No specific information about how attribution was measure was found. Not surprisingly, among studies of attribution in domestic violence cases, police officers are one of the most frequently studied groups. Stewart and Maddren (1997) studied the impact of alcohol involvement and gender on officers’ attribution in family violence cases. One unique aspect of this study was an effort to include both males (n=51) and females (n=46). Officers were exposed to a vignette, in which relationship and gender of the victim and perpetrator, as well as use of alcohol, were manipulated. Officers were then asked to assign independent levels of blame to both the victim and the perpetrator on a scale of 0 to 10. They were also asked to indicate on a seven point scale their likelihood of charging the assailant with a crime. Based on the results, gender did not affect attribution of causality. Greater blame was attributed to males who hit females compared with those hitting other males. Police blamed drunk victims more than sober victims. Victims drinking also had significantly more effect on attribution than did assailant drinking. Officers also showed a tendency to blame female victims less than male victims, suggesting a greater likelihood of arrest of the assailant if the victim was female. In a wide scale study of public attitudes, Gentemann (1984) conducted a telephone survey of a stratified, randomly selected sample of 422 female participants in North Carolina. Respondents were first asked if there were circumstances that justified, once in a while, a man swearing at, slapping, punching with a fist, or kicking his wife. Specific contextual justifications were added in the next set of questions and respondents were asked if a man should have a right to beat his wife if she was flirting with another man or if she was having an affair. They also asked if a man had a right to beat a woman if she was drunk or if she nags. From the responses of the contextual items, a Beating Justification Index (BJI) was derived which was a dichotomy consisting of beatings being justified in one or more conditions or beatings not justified in any condition. Almost 19% (n=422) of women endorsed justification under at least one of the contextual conditions. When demographic characteristics were taken into consideration, low-income respondents and those with less education justified abuse at a significantly higher rate. Participants were then given four hypothetical causal variables and asked if they believed the variables explained the reason for abuse. Models of learned behavior, unemployment, and low household income were supported as reasons for abuse (Gentemann, 1984). Gentemann (1984) also included 5 items to measure sex role attitudes. She predicted that sex role attitudes have to do with the extent to which a respondent endorses statements about traditional, stereotypical roles ascribed to males or females based only on gender. Domains measured included women working outside the home, cooking chores, and family and political authority. The relationship between the BJI and the EGALSCALE showed that those with traditional sex role attitudes were more likely to justify battering than were those with egalitarian attitudes. A more thorough testing of the effects of sex role stereotyping on domestic violence culpability attributions was conducted by Willis, Hallinan and Melby (1996). The attitudes of 120 European American (Caucasian) undergraduate students were measured. Each participant was given a trial transcript containing information about the testimony of a woman regarding a domestic violence assault. Manipulations included the victim’s race, the perpetrator’s race, and their level of intimacy (e.g. married or acquaintances). Students were also given the TESR (full title not found), a 20 item scale scored on a seven point Likert scale that assess agreement with traditional sex roles and traits of women and men. Also included was a questionnaire which assessed the respondents internal and external attributions of causality. Although Willis et al. (1996) found a wide variety of differences, some of the central variations distinguishing those with traditional sex role attitudes from those with more egalitarian attitudes included traditionalists being more sympathetic toward the man and assigning a rating of less abuse than their egalitarian counterparts. Also, traditionalists rated the incident as less abusive and recommended a shorter sentence when the woman was African American. In contrast, when the woman was European American, the traditionalists tended to recommend a longer term sentence than the egalitarians. The traditionalists were also less likely to label the attack as abusive when the man was African American. Intimacy of the relationship affected attributions regardless of sex role stereotyping. Crossman, Stith and Bender (1990) also undertook an examination of sex role egalitarianism and battering. The sample consisted of 115 men from either a treatment program for male violence (n=44) or from an alcoholism treatment program (n=71). Respondents were given the Approval of Marital Violence Scale (AMV) which asks the respondent to rate behavioral descriptors, such as a husband hitting for infidelity, on a seven point scale. The anchors of the scale were “justified” and “unjustified” or “acceptable” and “unacceptable.” An internal reliability coefficient of .84 was reported for the AMV. The Conflict Tactics Scale (CTS) – Form N (Strauss, 1979) was also given to measure exposure to violence in the previous year. Anchors are “never” and “more than once a month” on a six point scale. A modified (shortened) version of The Sex Role Egalitarianism Scale (SRE) (Beere, King, Beere, & King, 1984) was also used, for which an internal consistency reliability coefficient of .89 was reported. The SRE is a measure of sex role attitudes as previously defined. A social desirability measure and an alcoholism screening test were also administered. Crossman et al. (1990) found that sex role egalitarianism was explained by use of severe violence and attitudes supporting marital violence (inversely). Also of interest was that use of what was labeled as minor violence, which the violence treatment group scored higher on, did not contribute significantly to the regression model for egalitarianism. Also, 60% of the men in treatment for violence were alcoholic according to the alcoholism screening test, indicating a possible relationship of alcoholism to wife battering. Finn (1986) also conducted research on the relationship between sex role attitudes and attitudes supporting marital violence with a sample of 300 college students. The Personal Opinion Scale (POS), consisting of two subscales, the Attitudes Toward Sex Role (ASR) and the Attitudes Toward Force in Marriage (AFM), were developed. Items are responded to on a 5 point Likert scale with anchors of “strongly agree” and “strongly disagree.” A factor analysis reported the ASR emerged with an eigenvalue of 4.13 and factor loadings ranging from .57 to .91 and the AFM with an eigenvalue of 3.20 and factor loadings ranging from .46 to .78, indicating that each factor is a strong unidimensional measure of the desired attitudes. The majority of college students in the sample held egalitarian attitudes and did not approve of the use of physical force in marriage. A statistically significant relationship was also found between endorsement of traditional sex role attitudes and approval of the use of physical violence. Finn (1986) described with more precision some of the differences that emerged: …male students are more likely than female students to reflect traditional sex role attitudes and to hold attitudes that support the use of force in marriage. With respect to race, while white students are more likely to hold traditional sex role stereotypes, there is no difference between whites and blacks in their attitudes toward the use of force in marriage (p. 238). Another study of undergraduate college student attitudes toward violence in relationships was conducted by Bryant and Spencer (2003) and focused predominately on violence in dating relationships. Major findings supported Finn (1986) in that males were more likely to victim blame, thus likely demonstrating what Finn (1986) had termed traditional sex role attitudes. Bryant and Spencer (2003) also found, however that 39% of all students responding to the survey reported having used either physical, emotional, or sexual violence in a dating relationship within the preceding 12 months. There were also other differences found in attitudes including older students (juniors and seniors) being more blaming of societal factors and fraternity and sorority members were more blaming of the perpetrator than other students. Not surprisingly, students who reported using any form of violence in their relationships were more likely to demonstrate higher levels of victim blame than non-violent students. A statewide telephone survey was conducted in New Jersey by Edwards (1987). Although the sampling technique was thoroughly detailed, the total number of participants was not mentioned. The study specific questionnaire consisted of 33 questions and was intended to capture general public opinion on domestic violence. An analysis of the results indicated that subjects were asked about exposure to domestic violence and attitudes regarding the criminality of certain acts. Subjects were also apparently asked about roles of the court and law enforcement, reasons battered women do not leave the situation, and effects of violence on children. Although Edwards (1987) failed to report any statistical properties (other than percentages), there were a number of findings cited. The major findings included a report of violence against women in 21% of the homes surveyed, and rape and physical abuse as problematic with other intimates (e.g. dates) reported by 18% of respondents. Subjects were most likely to view acts involving injury as criminal and there was a strong preference for judges to deal with domestic violence with counseling as opposed to incarceration. Additionally, over half of all respondents endorsed a statement regarding strict enforcement as an effective deterrent. Approximately 76% considered domestic violence not to be a private family problem and a majority conveyed a belief that a violent partner would be violent no matter who they were married to, indicating a low degree of victim blame. With regard to not leaving a violent situation, lack of alternatives was the most frequently cited reason, specifically not having any place to go. Edwards (1987) also asked about characteristics respondents perceived to be associated with violent people. The most commonly endorsed characteristics included alcoholism, growing up with violence in the family of origin, and those lacking self-confidence. Koski and Mangold (1988) also conducted a telephone survey to study gender effects on attitudes about family violence. One hundred seventy-six subjects, eighteen or older, were sampled from northwest Arkansas. Questions assessed four areas of interest: Causes and extent of family violence, acceptance of specific acts of interpersonal violence, willingness to report, and belief in domestic violence as a public or private problem. Koski and Mangold (1988) found that while most respondents expressed that family violence was a problem, females were more likely to respond that it was a serious problem. Both genders heavily endorsed alcohol, drugs, and financial stressors as major causal factors. Females were more likely to endorse violence on television and increased stress due to a faster pace of life as contributing factors. When analyzing specific acts, widespread disapproval of most of the violent descriptors was found. With regard to effectiveness of responses to domestic violence, both males and females responded that family and friends are effective resources while males tended to also place greater value on private resources such as clergy and counselors than females. Women, by contrast, were more willing to endorse public avenues such as hotlines and shelters. Another important concept related to attribution of violence against women is how women are perceived on a societal level. This concept is referred to as sex role attitudes and was initially measured by Spence and Helmreich (1972) with the Attitudes Toward Women Scale (AWS). They tapped domains such as perspectives on educational, vocational, academic, dating behavior, and sex and marriage. The original scale consisted of 55 items responded to on a four point scale with “agree” and “strongly disagree” as the anchors. Scores reflected the degree to which the respondent held liberal or traditional views. Later Spence, Helmreich and Stapp (1973) developed a shorter 25 item version of the scale. Correlations between the long and short version ranged from .956 to .969, indicating that the short version and the long version were measuring closely related constructs. Parry (1983) developed a 22 item British version of the AWS. Nelson (1988) also conducted investigation of reliability, validity, and cross-cultural comparisons of the AWS. Her study supported use of the short version for both American and British samples. The AWS has also been used successfully to measure observer attitudes and judgements (Hiller & Foddy, 1993). Briere (1987) used a shortened version of the AWS in his study of prediction of the likelihood of battering from attitudes and childhood experiences. He also used The Acceptance of Interpersonal Violence Scale (AIV) (Burt, 1980), and he created the Attitudes Toward Wife Abuse Scale (AWA), along with a self-report of likelihood of battering under certain conditions, and a measure of early exposure to violence. Approximately 79% of all males in the sample reported some likelihood of using violence against a future wife in at least one of five scenarios, with sexual infidelity being the most commonly endorsed. Additionally, the AWS, AWA, and AIV correlated with four of the five situational conditions. Three scales for measuring attribution of violence stand out as the most systematically developed, empirically validated, and theoretically based multidimensional instruments. The first of these was developed by Saunders et al. (1987). The Beliefs about Wife Beating (IBWB) is a 41 question instrument with 5 subscales including wife beating justification, wives gain from beating, help should be given, offender punishment, and offender responsibility. The IBWB was normed on 675 psychology students and correlated with the Rape Myth Acceptance Scale (RMA) (Burt, 1980). Construct validity was established in a second study using several diverse samples totaling over 600 participants. Subjects were given the IBWB and other scales including the AWS (Spence, Helmreich, & Stapp, 1973), the Sex Role Stereotyping measure (SRS) (Burt, 1980), and the Hostility Toward Women Scale (HTW) (Check, Malamuth, 1983). Social desirability was also measured. A third validation study involved measurement of known groups of batterers and battered woman advocates. Results of the three separate validation studies confirmed the reliability and validity of the IBWB. Another well developed tool for measuring attitudes about violence in general, also applicable to the area of domestic violence, is the Violence Attitudes Scale (VAS) (Jackson et al., 1994). The first standardization study was conducted on 308 subjects and utilized 80 items. After the initial factor analysis the scale was reduced to 38 items and cross-validated using 301 additional subjects. Five factors are measured by the VAS including, in order frequency endorsed, criminal sanctions and perpetrator consequences, societal and family values, perpetrator internal characteristics, gang related issues and ethnicity, and victim blame. The most widely used multidimensional measure of attribution related to battering is the Domestic Violence Blame Scale (DVBS) (Petretic-Jackson et al., 1994). The DVBS is a 23 item instrument developed from an initial pool of 32 items. Items are scored on a six point Likert scale with anchors of “Almost never” and “Almost always,” or “Strongly disagree” and “Strongly agree.” The initial validation study was conducted with 424 young adults. After initial item analysis, 23 items were retained and from those items came four meaningful independent factors of blame: societal blame, situational blame, perpetrator blame, and victim blame. Data and cross-validation studies are available on a variety of populations including psychologists, mental health professionals, and physicians. Both theoretical conceptions and mathematical validations for the instrument were presented. An outline of major findings and their relationship to existing literature is explored and possible applications were discussed (Petretic-Jackson, Sandberg, & Jackson, 1994). Conclusion A variety of approaches to measuring attribution have been utilized in the study of wife battering. One notable shortcoming has traditionally been a lack of empirical, theoretically based, multidimensional instruments for measuring the construct. Although such instruments have been developed and are available, modified versions or independently developed lists of questions are used more frequently. If the literature in this area is to experience growth there must be movement toward greater uniformity in measurement of the relative constructs. A lack of standardization of measurement has several detrimental effects. First, it is not possible to be sure that studies proposing to examine similar phenomenon are measuring the same, or even similar, constructs. Inconsistency in measurement also makes replication very difficult, particularly when the instruments are not reproduced in the articles. Additionally, when attribution measures are used as outcome measures, the effect of the program cannot be assessed if the efficacy of the instrument is unknown. The ability to compare results across studies for the purpose of baseline norms or cross-validation would also help strengthen confidence in the findings. Another cause for concern was the large number of extremely short instruments used (5-10 questions) that relied exclusively on one item for the measurement of key factors. Although most of the studies utilizing such instruments did not address issues of reliability or validity, there is a likelihood that the use of single items could lead to increased susceptibility from intervening variables. As with many areas of social research, much of the work in the area of attribution of violence has been conducted on samples with limited representativness of society as a whole. Studies tend to rely heavily on students, police officers, and social service workers. Although these samples are important given their high probability of contact with victims or perpetrators, the literature could benefit from future broadening of investigations to known batterer samples, battering victims, a greater examination of judicial and jury attributions, and those of the general public. A final recommendation is for further exploration of malleable variables impacting attribution of violence for the purpose of designing more effective intervention and educational programs. Appendix Summary of noted study findings
Bibliography Acock,
A. C., & Ireland, N. K. (1983). Attribution of blame in rape cases: The
impact of norm violation, gender, and sex-role attitude. Sex
Roles, 9(2), 179-193. Arias,
I. (2000). The role of perceptions and attributions in battered women's intentions to permanently end their violent
relationships. Cognitive Therapy & Research, 24(2), 201-215. Bachman,
R., & Saltzman, L. E. (1995). Violence against women: Estimates from the
redesigned survey. Bureau of Justice Statistics Special Report.
(Publication #NCJ-154348). Washington, D.C.: U.S. Department of Justice. Beere,
C., King, D., Beere, D., & King, L. (1984). The sex-role egalitarianism
scale: A measure of attitudes toward equality between the sexes. Sex Roles,
10(7-8), 563-567. Briere,
J. (1987). Predicting self-reported likelihood of battering: Attitudes and
childhood experiences. Journal of Research in Personality,
21, 61-69. Brodsky,
S. (1976). Sexual assault: Perspectives on prevention and assailants. In
Walker, M. & Brodsky, S. (Eds.), Sexual assault, the victim and the
rapist (pp. 1-8). Lexington, MA: Lexington Press. Bryant, S. A., & Spencer, G. A.
(2003). University
Students’ Attitudes About Attributing Blame in Domestic Violence. Journal of
Family Violence, 18(6), 369-376. Bent-Goodley, T. B. (2004).
Perceptions of domestic violence: A dialogue with African American women. Health & Social Work, 29(4), 307-317. Burke,
D. M., Ames, M. A., Etherington, R., & Pietsch, J. (1990). Effects of
victim's and defendant's physical attractiveness on the
perception of responsibility in an ambiguous domestic violence case. Journal of Family
Violence, 5(3), 199-207. Burt, M.
R. (1980). Cultural myths and supports for rape. Journal of Personality and
Social Psychology, 38(2), 217-230.
Chadwick, B. A., & Heaton, T. B. (Eds.). (1992). Statistical
handbook on the American family. Phoenix, AZ: Orxy Press. Check,
J. V. P., & Malamuth, N. M. (1983). Sex role stereotyping and reactions to
depictions of stranger versus
acquaintance rape. Journal of Personality and Social Psychology, 45(2),
344- 356.
Crossman, R. K., Stith, S. M., & Bender, M. M. (1990). Sex role
egalitarianism and marital violence. Sex Roles, 22(5/6), 293-304. Davis,
L. V. (1992). Attitudes toward wife abuse in a cross-cultural context: A
comparison of Colombian and American human services students. In
E. C. Vivano (Ed.), Intimate violence: Interdisciplinary perspectives. (pp.229-242). Washington,
D.C.: Hemisphere Publishing Corp. Davis,
L. V. (1984). Beliefs of service providers about abused women and abusing men. Social
Work, 29(3), 243-250. Davis,
L. V., & Carlson, B. E. (1981). Attitudes of service providers toward
domestic violence. Social Work Research and Abstracts, 17(4), 34-39. Davis,
L. V., Hagen, J. L., & Early, T. J. (1994). Social services for battered
women: Are they adequate, accessible, and appropriate? Social Work,
39(6), 695-704. Dutton,
M. A. (1999). Multidimensional assessment of woman battering. Psychology of
Women Quarterly, 23(1), 195-199. Edwards,
C. (1987). Public opinion on domestic violence a review of the New Jersey
survey. Response, 10(1), 6-9. Finn, J.
(1986). The relationship between sex role attitudes and attitudes supporting
marital violence. Sex Roles, 14(5/6), 235-244. Finn, M.
A., & Stalans, L. J. (1995). Police referrals to shelters and mental health
treatment: Examining their decisions in domestic violence
cases. Crime and Delinquency, 41(4), 467- 480. Frieze,
I. H. (1979). Perceptions of battered wives. In Frieze, I. H., Bar-Tal, D.,
Carroll, J. S. (Eds.). New approaches to social problems. (pp.
79-108). London: Jossey-Bass Publishers. Gelles,
R. J., & Loseke, D. R. (Eds.).(1993). Current controversies on family
violence. Newbury Park, CA: Sage Publications, Inc.
Gentemann, K. M. (1984). Wife beating: Attitudes of a non-clinical
population. Victimology, 9(1), 109-119.
Greenblat, C. S. (1985). "Don’t hit your wife…unless…":
Preliminary findings on normative support for the use of physical force by husbands. Victimology,
10(1-4), 221-241. Hillier,
L., & Foddy, M. (1993). The role of observer attitudes in judgements of
blame in cases of wife assault. Sex Roles, 29(9/10), 629-644. Home, A.
M. (1994). Attributing responsibility and assessing gravity in wife abuse
situations: A comparative study of police and social workers. Journal of
Social Service Research, 19(1/2), 67-84. Howard,
J. A. (1984{a}). Societal influences on attribution: Blaming some victims more
than others. Journal of Personality and Social
Psychology, 47(3), 494-505. Howard,
J. A. (1984{b}). The "normal" victim: The effects of gender
stereotypes on reactions to violence. Social Psychology Quarterly, 47(3),
270-281. Jackson,
T. L., Dienst, R. D., Efird, T. L., Mobley, B. D., Schroeder, D. A., Hout, A.
D., Montecillo, J. C., & LaBine, A. L. (1994). The
Violence Attitudes Scale (VAS). In VandeCreek, L., Knapp, S., Jackson, T. L. (Eds.). Innovations
in clinical practice: A source book, Vol. 13. (pp. 279-291).
Sarasota, FL, USA: Professional Resource Press/ Professional Resource Exchange, Inc. Jones,
E. E., Kanouse, D. E., Kelley, H. H., Nisbett, R. E., Valins, S., & Weiner,
B. (1972). Attribution: Perceiving the causes of behavior. Morristown,
NJ: General Learning Press. Kalmuss,
D. (1979). The attribution of responsibility in a wife-abuse context. Victimology,
4, 284-291. Kanekar,
S., & Vaz, L. (1988). Attribution of causal and moral responsibility to a
victim of rape. Applied psychology: An international review, 37(1), 35-49. Koski,
P. R., & Mangold, W. D. (1988). Gender effects in attitudes about family
violence. Journal of Family Violence, 3(3), 225-237. Kurz, D.
(1993). Physical assaults by husbands: A major social problem. In Gelles, R.
J., Loseke, D. R. (Eds.). Current controversies on family violence. (pp.
88-103). Newbury Park, CA: Sage Publications, Inc. Lamb,
S., & Keon, S. (1995). Blaming the perpetrator: Language that distorts
reality in newspaper articles on men battering women. Psychology of Women
Quarterly, 19, 209-220. Lavoie,
F., Joacob, M., Hardy, J., & Martin, G. (1989). Police attitudes in
assigning responsibility for wife abuse. Journal of Family Violence, 4(4),
369-388.
Luginbuhl, J., & Mullin, C. (1981). Rape and responsibility : How
and how much is the victim blamed. Sex Roles, 7(5), 547-559. Nelson,
M. C. (1988). Reliability, validity, and cross-cultural comparisons for the
simplified attitudes toward women scale. Sex Roles, 18(5/6),
289-296. O'Brien,
K. M., & Murdock, N. L. (1993). Shelter Workers' Perceptions of battered
women. Sex Roles, 29(3/4), 183-194. Paquin,
G. W. (1994). A statewide survey of reactions to neighbors' domestic violence. Journal
of Interpersonal Violence, 9(4), 493-502. Parry,
G. (1983). A British version of the Attitudes Toward Women Scale (AWS-B).
British Journal of Social Psychology, 22, 261-263. Petretic-Jackson, P., Sandberg,
G., & Jackson, T. L. (1994). The domestic violence blame scale (DVBS). In
VandeCreek, L., Knapp, S., Jackson, T. L. (Eds.). Innovations in
clinical practice: A source book, Vol.
13. (pp. 265-278). Sarasota, FL, USA: Professional Resource Press/ Professional
Resource Exchange, Inc.
Saunders, D. G., Lynch, A. B., Grayson, M., & Linz, D. (1987). The
inventory of beliefs about wife beating: The construction and initial
validation of a measure of beliefs and attitudes. Violence and Victims, 2(1),
39-57.
Schwartz, M. D., & Mattley, C. L. (1993). The battered woman scale
and gender identities. Journal of Family Violence, 8(3), 277-287. Sigler,
R. T. (1989). Domestic violence in context: An assessment of community
attitudes. Lexington, MA: Lexington Books. Spence,
J. T., & Helmreich, R. (1972). The Attitude Toward Women Scale: An
objective instrument to measure attitudes toward the rights and roles of women
in contemporary society. JSAS Catalog of Selected Documents in Psychology, 2,
66. Spence,
J. T., Helmreich, R., & Stapp, J. (1973). A short version of the Attitudes
Toward Women Scale (AWS). Bulletin
of the Psychonomic Society, 2(4), 219-220. Stalans,
L. J., & Finn, M. A. (1995). How novice and experienced officers interpret wife
assaults: Normative and efficiency frames. Law and Society Review, 29(2),
287-321. Stewart,
A., & Maddren, K. (1997). Police officers' judgements of blame in family
violence: The impact of gender and alcohol. Sex Roles, 37 (11/12),
921-933. Strauss, M. A. (1979). Measuring intrafamily
conflict and violence: The Conflict Tactics Scale. Journal of Marriage and
the Family, 41, 75-88. Straus,
M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors:
Violence in the American Family. Garden City, NY: Anchor Books.
Tewksbury, R., & Adkins, M. (1991). Rape myths and emergency room
personnel. Response, 14(4), 10-15.
Tucker-Ladd, C. E. (1996). Psychological Self-Help. Dublin, Ohio: Mental Health Net. Waaland,
P., & Keeley, S. (1985). Police decision making in wife abuse: The impact
of legal and extralegal factors. Law and Human Behavior, 9(4), 355-366.
Williams, O. (1990). Spouse Abuse: social learning, attribution and
interventions. Journal of Health and Social Psychology, 1(2), 91-106. Willis,
C. E., Hallinan, M. N., & Melby, J. (1996). Effects of sex role
stereotyping among European American students on domestic violence culpability attributions. Sex
Roles, 34(7/8), 475-491. |