Table of contents
Prevalence and Public Health of IPV
Gender- Partner Violence Implicit Association Test
Associations Between the IAT and Explicit Measure
Correlations among Explicit Measures
Research on attitudes towards Intimate Partner Violence (IPV) has mainly focused on explicit attitudes, while partner violence may have automatic behavioral component guided by implicit beliefs and attitudes. The present study aimed to investigate if the implicit attitude towards IPV varies systematically as a function of observers and perpetrators gender, 94 young adults (20 men and 74 women) participated in an Internet-based study and took a version of the Implicit Association Test (IAT), an implicit measure of attitudes towards gender-specific partner violence. Participants completed explicit measures which assessed partner violence victimization and perpetration, emotional flooding, rape myth acceptance, explicit attitudes towards IPV and depressive symptoms. The analysis revealed that women strongly preferred female-perpetrated IPV over male-perpetrated IPV, while male participants showed no such bias. This finding provides evidence for the existence of gender-related differences in the implicit attitude towards violence within intimate relationships.
Keywords: gender; bias; attitudes; intimate partner violence; IPV; Implicit Association Test; IAT; emotional flooding; violence; health;
Explicit and Implicit attitudes towards the justifiability of partner violence are two important factors which influence the perception of IPV-events and the attribution of blame and responsibility. Implicit attitudes towards IPV have not received much attention in the literature , but the investigation of implicit attitudes could lead to a more refined theory about the etiology of partner violence. Moreover, implicit attitudes can influence the reaction of society (e.g. friends, service providers, health professionals) towards victims of partner violence. The present study aimed to examine gender-related differences in the implicit attitude towards partner violence.
Prevalence and Public Health of IPV
Violence is described by the WHO as a public health problem, the prevention of violence within intimate relationships is a vital public health goal. Intimate Partner Violence (IPV) has long been synonymous with wife battering and forms of partner violence which have their basis in the dynamics of power and control. A pioneering study from Makepeace(1981) shed light on a different form of partner violence has been labeled by Johnson as Situational Couple Violence (Kelly, J. B., & Johnson, M. P., 2008). This form of IPV is a prevalent problem particularly among younger, dating, and student samples and is caused by everyday frustrations and stressors of living together (Archer, 2000, 2002). There has long been a controversy on weather the majority of violence in relationships is perpetrated by men. A series of studies have shown that partner violence most often is mutual and that women and men can be both perpetrators and victims of IPV (Archer, 2000; Shook, N. J., Gerrity, D. A., Jurich, J., & Segrist, A. E., 2000; Straus, 2004; Straus & Ramirez, 2007b; Whitaker, Haileyesus, Swahn, & Saltzman, 2007). Prevalence of IPV varies depending on several factors including the region, sample characteristics and that are used for assessment (Langhinrichsen-Rohling, Selwyn, & Rohling, 2012; Straus, 2004; Straus & Ramirez, 2007b).
A review about physical couple violence among college students found prevalences ranging from 16.7 % to 48 % within the last 12 months (Murray, C. E., & Kardatzke, K. N.). Most studies conducted in North America report prevalences ranging from 21% to 29 % (Amar, A. F., & Gennaro, S., 2005; Straus & Ramirez, 2007b; Whitaker et al., 2007), the prevalence is similar in Europe. Moreover, two studies from Germany and Sweden found that males reported a higher frequency of physical IPV victimization than females, although mostly for minor forms (Danielsson, Blom, Nilses, Heimer, & Högberg, 2009; Krah & Berger, 2005). Similarly, a study showed that in the 50% of cases where physical violence is not reciprocal, women were the perpetrators in more than 70% of the cases (Whitaker et al., 2007). Psychological aggression is even higher with a prevalence of approximately 80 to 90 % within the last 12 months.
Intimate partner violence has shown to be related to serious health problems. Several studies reported adverse effects of partner violence on women`s mental health, victims have higher rates of depression, chronic pain and PTSD (Campbell, 2002; Golding, 1999; Johnson, Giordano, Longmore, & Manning, 2014; Pico-Alfonso, M. A., Garcia-Linares, M. I., Celda-Navarro, N., Blasco-Ros, C., Echeburúa, E., & Martinez, M., 2006; Sugg, 2015). Moreover, partner violence victimization is associated with chronic diseases of the cardiovascular system and asthma (Sugg, 2015). Furthermore, both physical and psychological partner aggression is related to physical and mental health consequences for female and male victims (Coker et al., 2002). Finally, a study among students showed that individuals who experience multiple forms of partner aggression have significantly more problems with mental health and also reported more injuries (Amar, A. F., & Gennaro, S., 2005). These findings demonstrate that violence within intimate relationships is a significant public health problem, and attitudinal acceptance of IPV is one of its most important aspects.
Attitudes are beside stereotypes one of the oldest and most-studied concepts in the analysis of social behavior (Brown, R., & Gaertner, S, 2008; Greenwald, A. G., & Banaji, M. R., 1995). While stereotypes guide behavior towards a member of a group as if the other possesses traits included in the group stereotype, attitudes are favorable and unfavorable dispositions towards an object of thought, e.g. things, ideas, people, groups, policies or places (Bohner, G., & Dickel, N., 2011; Greenwald, A. G., & Banaji, M. R., 1995). Researcher measure attitudes and beliefs not only with the goal to them, but they are also interested in the predictive validity of attitudes for an individual`s future behavior. Meta-analytic research has found a moderately strong attitude-behavior consistency, although the strength of the relationship between attitudes and behavior shows considerable variability across different study topics (Glasman & Albaraccin, 2006; Greenwald, Poehlman, Uhlmann, & Banaji, 2009; Kraus, 1995). Furthermore, attitudes can be measured directly or indirectly, contemporary theories of attitude also distinguish between explicit and implicit attitudes. Direct methods are widely used in attitude research and make it obvious that respondents self-reported opinion is requested. Explicit self -report scales were until recently by far the most popular instruments to measure attitudes (Gawronski & Bodenhausen, 2006; Wilson, T. D., Lindsey, S., & Schooler, T. Y., 2000).
Assessing explicit attitudes. The investigation of attitudes towards intimate partner violence is very popular among IPV-researchers. (Slep, Cascardi, Avery-Leaf, & O'Leary, 2001). Studies using methods like self-report and vignettes found that violent behavior within intimate relationships is associated with accepting attitudes (Hanson, R. K., Cadsky, O., Harris, A., & Lalonde, C., 1997; Rhatigan, Stewart, & Moore, 2011; Stith & Farley, 1993). Consequently, scholars argue that views which justify the use of aggression against intimate partners are an important risk factor in the etiology of IPV. Furthermore, individual and collective perceptions of IPV- events play a crucial role in the response from institutions. Moreover, most efforts for prevention and intervention of IPV focus for practical reasons directly on changing attitudes and beliefs, especially through educational programs and public media campaigns (Eigenberg & Policastro, 2015; Slep et al., 2001).
Attitudinal research on IPV found that attitudes and perspectives about IPV vary depending on sociodemographic characteristics, victim behavior before the attack and prior IPV perpetration. Both observers and perpetrators gender influence the perception of an IPV- event, and with it also the justifications for the violent behavior and the assignment of blame (Rhatigan et al., 2011; Simon, T. R., Anderson, M., Thompson, M. P., Crosby, A. E., Shelley, G., & Sacks, J. J., 2001; Stewart, Moore, Crone, Defreitas, & Rhatigan, 2012). Research which focused on the justifiability of male`s use of violence against their female partners found that men hold more accepting attitudes towards intimate partner violence compared to women (Coughlin, 1994; Dutton & Nicholls, 2005; Wang, 2016; Witte, T. H., Schroeder, D. A., & Lohr, J. M., 2006). Men were also found to hold slightly more accepting views of IPV compared to women in a study that also assessed participants attitude towards partner violence perpetrated by women. Although overall woman endorsed attitudes towards IPV that were less accepting or justifying of partner violence, it is possible that type of behavior – and not the gender of the perpetrator - caused the difference in attitudinal acceptance between male and female participants (Slep et al., 2001). However, if perpetrators gender plays a major role in the attitude towards the justifiability of partner violence then there is not just one concept of "attitude towards the justifiability of IPV," but two different concepts for the attitude towards the justifiability of male and female aggression against their partner (Slep, 2001). Studies which assessed attitudes with different items for men and women and described specific conflict situations found stronger associations between aggression and behavior, particularly for women (Foo & Margolin, 1995). Moreover, attitudinal acceptance of women towards female aggression was found to be related to female partner violence, and attitudinal acceptance of men towards the use of violence against female partners was related to male partner violence (Foo & Margolin, 1995; Slep et al., 2001). Congruent with the theory of two different concepts of attitudinal acceptance for male and female IPV perpetration, as series of studies has shown that the gender of the perpetrator influences the perception of aggressive behavior within intimate relationships, i.e. if a certain action is considered normative or transgressive by observers.
Research indicates that male and female observers attribute less blame and responsibility to women who attack their partners than to men who attack their partners, for similar acts. In particular female observers attribute more blame to males who use aggressive behavior towards their partners. Besides, male perpetration is perceived as more criminal (Ferguson & Negy, 2004; Rhatigan et al., 2011; Simon, T. R., Anderson, M., Thompson, M. P., Crosby, A. E., Shelley, G., & Sacks, J. J., 2001; Taylor & Sorenson, 2005).This gendered bias extends to situations where a perpetrator hits a verbally or physically confrontational victim. More blame is attributed to a male who gets violent towards a confrontational partner than to a woman who gets violent towards her confrontational or even unconfrontational partner. Furthermore, college students perceived the aggressive behavior of male perpetrators as more stable and more likely caused by negative personality characteristics while the violent behavior of females is seen as more infrequent and as a response to situational events (Rhatigan et al., 2011). Situational events like confrontational behavior have been investigated by a series of studies and play a role in the attitude towards the justifiability of partner violence.
Despite this negative evaluation of male-perpetrated IPV, a series of studies show that males view partner violence as more justifiable than women if it is in response to confrontational and aggressive behavior. Simon et al.(2001) found that men find hitting a partner after being hit more acceptable, regardless of the gender of the hitter. However, and there is no evidence that attitudes which justify self-defense contribute to IPV-perpetration. For example, research examining the attitudes of college students found that approval of self -defense is not associated with self-reported intimate partner violence, in contrast to reporting that aggression in response to humiliation is acceptable (Foo & Margolin, 1995). Moreover, Witte et.al (2006) found men attributed more blame to victims than females if the victim was verbally aggressive before the attack. However, this study only focused on male-perpetrated violence against female partners and did not assess participants attitude towards violence from women against a verbally aggressive male partner; this leaves open the possibility of own- gender bias as the organizing force behind the findings of this study.
Besides gender and context factors, prior IPV-perpetration has been found to play a role in the evaluation of intimate partner violence. Perpetrators of past psychological and physical abuse tend to attribute less blame to fellow perpetrators than individuals without experience in IPV-perpetration. Rhatigan (2011) argues that this tendency to put less blame on similar others is a case of in- group/out-group bias and can explain the bias of perpetrators as well as the gendered bias of male and female observers. Another theoretical explanation for the gendered bias against male aggression argues that the general public views women as less aggressive than men. According to this theory, people unintentionally use stereotypes about female and male traits when they interpret IPV-events (Straus & Ramirez, 2007a; Winter, L., & Uleman, J. S., 1984). These explanations both suggest that implicit processes play a major role in the perception of IPV-events.
Assessing implicit attitudes. Attitudes can not only be measured with direct but also with indirect tasks (Gawronski, B., & Bodenhausen, G. V., 2007; Moors & DeHouwer, 2006; Petty, R. E., Fazio, R. H., & Briñol, P., 2012). Indirect methods have a long tradition in psychology and were used when the direct method (i.e. asking people about their opinions) was not possible for a certain research question or would have led to data of limited value. Indirect measures have advantages over self- report measures, self -report measures rely on respondents willingness and ability to report private knowledge (Greenwald et al., 2002).
In recent years a series of new implicit measures have been developed, for an overview see Bar-Anan and Nosek (2014). The most popular of this new measures probably is the Implicit Association Test (IAT; (Greenwald, A. G., McGhee, D. E., & Schwartz, J. L., 1998b). The IAT is a categorization task that has been designed to assess individual differences in implicit cognition. In different within-subject conditions, subjects respond to two categories of target stimuli and two categories of attribute stimuli. The categorization of these stimuli is easier and thus faster when closely related items share the same response key (Greenwald, A. G., McGhee, D. E., & Schwartz, J. L., 1998b; Nosek, Greenwald, & Banaji, 2005). The IAT can be easily adapted to any target or attribute stimuli. The method is especially useful to investigate attitudes about sensitive topics, i.e., where explicit measures would be influenced by social desirability.
Independent from the method of investigation attitude theorists distinguish between explicit and implicit attitudes. Explicit attitudes are related to deliberate behavioral intentions, whereas “automatic” implicit attitudes predict spontaneous behavior (Gawronski & Bodenhausen, 2006; Rydell & McConnell, 2006). Models which include explicit and implicit attitudes are referred to as dual process models and are utilized in various psychological disciplines. For example, the model is used to examine and describe health behavior. The dual process model of health behavior differentiates between "cold" reflective processes (e.g. avoiding threats to long-term health) and “hot” implicit processes that trigger approach towards a temptation despite health risks (Wiers, R. W., Houben, K., Roefs, A., De Jong, P. J., Hofmann, W. & Stacy, A. W., 2011).
Research using the dual process theory found that automatic attitudes towards a temptation have a stronger influence on behavior if a person has lower working memory capacity or lower self-regulatory resources. Conversely, explicit attitudes have shown to guide the behavior of people with a high working memory capacity (Finkel, DeWall, Slotter, Oaten, & Foshee, 2009; Hofmann, Gschwendner, Friese, Wiers, & Schmitt, 2008). For example, one study found that self -regulatory failure leads people to act on their gut instinct. Although individuals with higher levels of self-regulatory resources also experience violent impulses during conflicts within intimate relationships, they have the ability to override these impulses. Research also showed that the experimental depletion of self-regulatory resources increased violent behavior against the partner. Conversely, experimentally bolstered self-regulatory resources decreased violent inclination (Finkel et al., 2009).
Studies which investigated the implicit associations of male IPV perpetrators found that they did not evaluate women more negatively than men on a gender evaluation IAT. Moreover, the study concluded that they did associate women with violence and had a more positive evaluation of violence, especially if they experienced more verbal aggression from their partner. Also, a marginally significant relationship emerged between a positive evaluation of violence and IPV victimization and perpetration (Eckhardt & Crane, 2014; Eckhardt, Samper, Suhr, & Holtzworth-Munroe, 2012). Moreover, the finding from this study with IPV-perpetrators that men do not evaluate woman more negatively is consistent with recent studies which investigated the development of implicit own-gender preference (Dunham, Baron, & Banaji, 2016).
A series of IAT- studies found that boys and girls preferred their respective gender (Cvencek, Greenwald, & Meltzoff, 2011; Dunham et al., 2016). However, as males grow up the preference steadily decreases until it reaches a neutral level in adulthood. Conversely, research shows that females own-gender preference stays at a high level. As possible explanations for their findings, the authors discuss people`s relationship to their primary caregiver (i.e. the mother) and increased competition among males. They focus on the anomaly that the supposedly higher status group is evaluated more negatively and notice that the results suggest that women`s discrimination is not caused by aversion against women. However, among other things, such a bias could influence the attitude towards the justifiability of partner violence and the attribution of blame and responsibility.
Research suggests that explicit attitudes towards the justifiability of partner violence are influenced by the gender of the perpetrator and the observer. Less is known about the implicit attitudes towards IPV. The investigation of implicit attitudes towards IPV can contribute to the development of a new refined theory of the etiology and attitude towards IPV-events. The present work addresses the following questions:
(1) Do the implicit attitudes towards physical IPV, as measured through an Implicit Association Test (IAT), vary systematically as a function of observers and perpetrators gender?
(2) Which critical variables can be identified that correlate with implicit and explicit attitudes towards IPV?
If there is consistency between explicit and implicit attitudes, male-perpetrated violence should also be evaluated more negatively when measured with an implicit method. The relationship between explicit and implicit attitudes and the relationship between those attitudes and IPV-related variables is another goal of this study. The variables to be examined will include IPV- perpetration and victimization, emotional flooding, depressive symptoms, rape myth acceptance, and relationship satisfaction.
Ninety-four adolescents and younger adults (20 men, 74 women) participated in the Internet-based study. The age-span ranged from 20 to 39 years (M = 25.9, SD = 4.23). The average age of female participants in the sample was 25,3 (SD = 3.9). The average age of male participants in the sample was 28 (SD = 4.8). Most participants had a completed degree (63.6%) or a general qualification for university entrance (34.1%). Moreover, most participants classified themselves as students (72.5%). Over 90% of participants reported being either German (87.4%) or Austrian (7.4%) citizens. Furthermore, most participants had been in an intimate relationship, with a duration of at least one month, within the last 12 months (67%) or before (22.4%). The average relationship length was 8.1 years (SD = 1.6 years)
From the visitors who visited the study website, 116 proceeded beyond the welcome page. Most of these people completed the study (78.4%), approximately one-fifth (19.5%, N = 17) left the study either on the first page (participant demographics) or the second page (Implicit Association Task, IAT).
The internet-based study was created with SoSci Survey (Leiner, 2014), a software which is free for scientific surveys. A chargeable add-on module was booked which allows researchers to combine their questionnaires with implicit methods like the Implicit Association Test (IAT). After a pretest phase, the research project was published and was available for participants for two weeks between 10. June 2017 and 23. June 2017. To recruit participants adverts that gave a brief description of the study and provided a link to the study webpage were posted on different sites related to science communication and research. Visitors of the study website were thanked for their interest in the study and could read some information about the research project. They were informed that a desktop PC/ Laptop would be required to take part and that they could enter a raffle for two € 25 Amazon vouchers for their participation. Informed consent was obtained before visitors proceeded to the first page of the study where they were asked about their sociodemographic profile. Participants then took the attitudes towards gender-specific Intimate Partner Violence IAT; the classification task took participants approximately 5 minutes. After they had finished the IAT, participants then completed questionnaires assessing relationship experience, aggressive conflict tactics in the relationship, emotional flooding symptoms, gender identity, endorsement of feminist ideology, acceptance of gender-specific IPV, rape myth acceptance and depressive symptoms. Finally, participants were thanked for their contribution.