Just World Responses Towards Victims of Child Sexual and Physical Abuse
Frequently, individuals in their day to day lives are exposed to the unjust suffering and the victimization of innocent others. How individuals respond towards innocent victims has been the subject of much research. Many research findings have demonstrated that there is often a tendency to blame the victim for their own misfortune. One theory that has been formulated to explain this phenomenon is Just World Theory (JWT) which states that individuals are motivated to believe that the world is a just place in which people get what they deserve. When confronted with the presence of an innocent victim, individuals find reasons to blame the victim for their misfortune to minimalize the threat that the victim poses to their belief in a just world.  The phenomenon of victim blame has also been explored by researchers investigating various responses to child victims of sexual abuse. However, this has largely been independent of the theoretical perspective of JWT. Therefore, developments in just world theory, such as the presence of multiple co-occurring responses towards victims other than blame, often have not been taken up by researchers investigating responses towards child victims of sexual abuse. The current study attempts to explore responses made towards child victims of both physical and sexual abuse using the theoretical framework of JWT. It allows for the presence of multiple responses as well as individual difference variables that may predict the use of certain responses. Fake news articles were used as stimuli to increase stimulus impact instead of commonly used, hypothetical vignette scenarios.
The social response to trauma
Human reactions to trauma and victimization highlight the precarious nature of social schemata designed to enable a coherent mental framework from conflicting perceptions, memories and events. Survivors of significant trauma often struggle to make sense of their tragedy and attempt to cope with a ‘shattered worldview’ characterized by a sense of incomprehensibility and a loss of meaning in the world (Janoff-Bulman, 1992). The attempt to make sense of a traumatic experience is primarily concerned with the distribution of negative outcomes and is exemplified by not simply asking “Why?”, but by asking, “Why me?” (Janoff-Bulman & Yopyk, 2004). To the survivor, traumatic events do not make sense to their fundamental ‘assumptions’ or ‘schemata’ about the world, which are taken for granted as implicit ‘rules’ or ‘laws’ governing notions of outcome distribution and deservedness. These fundamental or basic assumptions are characterized by implicit beliefs about the world in terms of benevolence, meaningfulness and individual self-worth (Janoff-Bulman, 1989; 1992). Individuals are strongly motivated to believe that the world is good, orderly, predictable and just (Janoff-Bulman & Yopyk, 2004). Believing the world is random, unpredictable and uncontrollable makes safety and security less assured. Essentially, fundamental assumptions are a kind of necessary illusion. Beliefs in the inherent ‘goodness’ of the world, justice and an inflated sense of one’s control over events, have positive benefits for mental health and well-being in that they are conducive to cognitions that enable individuals to carry out the functions of daily life (Furnham, 2003).
In much the same way that a survivor of trauma is forced to radically alter their fundamental assumptions about the world, the presence of an individual who has undergone a traumatic experience poses a unique kind of threat to the worldview of the wider social group which they inhabit. This is reflected in societal attitudes which seek to de-emphasise information that is fundamentally incongruent with basic assumptions of meaning, goodness and behavioural control over outcomes. The presence of trauma seems to pose such a challenge to the public imagination that there is a kind of shared amnesia to the suffering and helplessness that traumatic experiences such as child abuse elicit (McFarlane, 2000). Historically this has been present in attitudes of the medical community, military, government officials, and the wider public towards soldiers and veterans suffering psychiatric disorders relating to combat exposure. This is readily evident in accounts of individuals during the First World War who faced public accusations of cowardice, medical statements of ‘moral inferiority’, court martial and even the firing squad, for psychiatric conditions clearly relating to combat exposure (McFarlane, 2000). Such accusations and denial of the impact of trauma appear to be social mechanisms for maintaining the necessary illusion of the fundamental assumptions. In a sense the presence of a victim of trauma is a glaring reminder of one’s own vulnerability to harm, lack of control over outcomes, and is a threat to the fundamental sense of meaning and goodness in the world.
The Justice Motive
Various sources have noted the tendency for people to respond negatively towards a victim of trauma, often taking the form of perceived behavioural responsibility of a victim for their misfortunes.  Blaming the victim of trauma may serve as a defence mechanism unwanted appraisal of basic assumptions. Theorists postulate that blaming the victim of trauma is related to a global justice motive (Lerner & Miller, 1978; Hafer and Bègue, 2005; Furnham, 2003). This justice motive often takes the form of personal beliefs that the world is a just place and that people get what they deserve. The belief in a just world (BJW), enables individuals to confront their physical and social environment as if it were stable and orderly (Lerner & Miller, 1978). In a sense, this belief serves an important adaptive function in an individual, facilitating the socially regulated behaviours of day-to-day life. This is essentially similar to the theoretical basis of terror management whereby the pursuit of self-esteem and investment in a cultural worldview serves as a buffer against existential anxiety and allows individuals to engage in the automatic activities of day-to-day life (Janoff-Bulman & Yoprk, 2004). Thus, the belief in a just world and investment in a cultural worldview serve to create stability, order and meaning in an otherwise arbitrary and threatening world in which individuals have little control over events, are vulnerable to harm and extinction and where people often do not get what they deserve. Due to this, individuals share a perceived interdependence between their own fate and that of others, and are thus motivated to restore justice by either acting to compensate the victim or by persuading themselves that in some sense the victim deserved it (Lerner & Miller, 1978).
The just world hypothesis was originally postulated by Melvin Lerner following a series of experiments. The first of which involved participants observing two students undertaking a problem solving task in which rewards were allocated randomly via a lottery. He found that people tended to believe that the student who had been rewarded had worked harder at the task than the student who was not rewarded. Lerner theorized that people tend to interpret social events in a manner that fits into a simple and understandable process (Lerner, 1965). Lerner speculated that this process is often guided by basic notions of deservingness as a means in which to find a fit between a person’s character and the outcomes that they receive. JWT in its subsequent form was conceived of due to the results of a ground-breaking experiment in which participants witnessed via videotape a confederate ‘victim’ reacting with apparent pain to a series of supposed electric shocks as a course of punishment for errors in a supposed learning task (Lerner & Simmons, 1966). Lerner and Simmons found that subjects who had observed the victim being electrocuted tended to derogate the character of the victim when they were informed that it was to continue in another session or that the victim had volunteered to submit themselves to a second session for the sake of the study.  However if they were given the option to place the victim in another condition where they would be not be shocked and would instead be rewarded, participants were far less likely to derogate the character of the victim. Lerner and Simmons (1966) argued that the sight of an innocent victim which they were powerless to help, suffering without possibility of reward or compensation motivated people to devalue the victim in order to bring about a ‘fit’ between the victims fate and their character so as to maintain the participants belief in a just world. Subsequent research supports just world theory (Furnham, 2003; Furnham & Proctor, 1989; Hafer & Bègue, 2005; Lerner, 1980; Lerner & Miller, 1978).
Theorists have postulated that the belief in a just world is only one part of a more general global justice motive (Furnham, 2003; Hafer & Bègue, 2005; Lerner & Miller, 1978). This general justice motive is evident given that justice concerns are inextricably bound up in the complex process of how individuals perceive, define, evaluate and respond to what is happening to themselves and others (Lerner, 1981). Hafer and Bègue (2005) note that the most well specified reason for why people develop a general justice motive is to maintain their social contract. Lerner, (1977) proposed that the development of the social contract is a natural product of a child’s progression from the immediate gratification of impulses to investment in delayed gratification for long-term and better outcomes. The social contract teaches the child that the world is a place where their investment will entitle them to desired outcomes and that earning or deserving is an effective way of obtaining desires. Thus, as a child ages they develop an increasing sensitivity to deservingness and as greater portions of their goal seeking activities are placed under the rules of deservingness, eventually their life is committed and organized on the basis of deserving outcomes. Hafer and Bègue, (2005) write that the fate of others becomes a central concern because indications that others are not getting what they deserve threatens the notion that the world is a place in which people earn their outcomes which in turn threatens an individual’s commitment to their own personal contract. Support for this is present in early work by Braband and Lerner (1974) and Long and Lerner (1974) that demonstrated an association between the development of children’s delay of gratification and an increasing concern with deservingness. Furthermore, subsequent research by Hafer (2000b) has demonstrated in adults that there is a strong relationship between investment in long-term goals and the personal belief in a just world.
Individual Differences
Although Lerner had originally conceived of the belief in a just world as a ‘useful metaphor’ rather than a clear psychological construct (Lerner, 1997), psychometric scales measuring individual differences in the need to believe in a just world have been developed and are widely used in contemporary belief in a just world (BJW) research. Currently predominate in the literature, Rubin and Peplau’s (1973, 1975) Just World Scale is the most widely used scale, despite factor analyses finding it to be multi’ rather than uni-dimensional, as well as the presence of alternative scales such as Lipkus’s Global Belief in a Just World Scale that has superior psychometric properties (Hellman, 2008; Lipkus, 1991). There has also been strong criticism of the use of individual difference scales in the literature. Hafer and Bègue (2005) argue that although newer BJW scales are more reliable they still contain measurement and conceptual issues. For example, they note that although the relationship has been inconsistent, several studies have found significant relationships between social desirability and scores on a BJW scale. Further research by Alves and Correia (2008, 2010) has found that the expression of higher BJW is considered to be more normative and socially valued than the expression of low BJW. Another criticism of the use of BJW scales revolves around the issue of construct validity, especially considering that Lerner has continually suggested that the belief in a just world is implicit and is very rarely explicitly endorsed, yet the most widely used means of measuring it is with an explicit measure. Lastly, Schmitt (1998) reports that there is a relative skew in scores on BJW scales towards the low end of responding, backing up Lerner’s claim that BJW is rarely explicitly endorsed.
HERE
	Regardless of these criticisms, studies involving experimental manipulation have continually demonstrated that individuals appear to respond to injustice as if they did believe in a just world. The crux of the matter is that just world scales are so prevalent in the literature that often studies will require a significant interaction between participants scores on a just world scale and experimental manipulations for any evidence of just world based mechanisms (Hafer & Bègue, 2005). This problem has been highlighted in a study by Maes (1998) who explored immanent and ultimate forms of justice beliefs. Maes found that both immanent justice beliefs, (the belief that everything that happens is the result of previous misdeeds); and ultimate justice beliefs, (the belief that all injustice is reconciled in the long run by future justice) were significantly related to scores on a scale designed to measure the belief in a just world. However, despite this, immanent and ultimate forms of justice differed significantly in their relationships to certain responses towards a hypothetical victim. Interestingly, Maes found that strong immanent justice beliefs were significantly associated with the frequently reported blame and derogation of victims in the JWT literature. However conversely,  high levels of belief in ultimate justice were significantly associated more with positive rather than negative appraisals of the victim. Results like this indicate that scales designed to measure individual differences in the belief in a just world are unable to measure the complexities of an individual’s justice related beliefs. The prevalent use of just world scales in their current manner can only lead to further inconsistency in proposed predictions between individual differences and experimental manipulations and at best, sporadic support for hypothesized just world effects (Hafer & Bègue, 2005).
However, despite these criticisms, although BJW individual difference scales can’t be relied upon to reveal the totality of an individual’s justice related beliefs and attitudes, they can be useful in that they are capable indicators of some of the explicit justice beliefs that people develop during the course of their lives in response to injustice (Hafer & Bègue, 2005). Interestingly, several authors have suggested that there may be implicit and explicit versions of the belief in a just world (Dalbert, 2009).  Furthermore highlighting this complication with the widespread use of scales in the JWT literature, Lerner (1998) in his reflection on JWT proposed that there were two separate ways of conceptualizing BJW. The first conceptualization is conscious and relates to conventional rules, morality and social judgements. The other is preconscious and includes primitive rules of blaming and automatic emotional consequences (Lerner, 1998). These two conceptions of BJW echo the experiential versus rational distinctions that are employed by individuals in their day-to-day evaluation of environmental information (Epstein, Lipson, Hostein & Hub, 1992; Lerner, 1998, 2003). The conscious conception of belief in a just world as represented by the use of psychometric scales such as Rubin and Peplau’s (1973, 1975) Just World Scale have generally focussed on correlational relationships between participants self report scores on belief in a just world scales and various other criteria such as attitudinal and personality variables and measures of well-being (For reviews, see Furnham, 2003; Furnham & Procter, 1989). The preconscious, more traditional conception of the belief in a just world, that BJW is an implicit motive that is present in social responses towards victims that do not follow the rules of rational assignment of blame and tend to eliminate or reduce the element of unjust or undeserved suffering (Lerner, 1998). This conception of BJW is most generally explored using experimental means, for example presenting participants with a victimization scenario and manipulating variables (e.g., victim innocence) that are expected to increase or decrease the threat that the outcome of the scenario poses to the participants need to believe in a just world (for review, see Hafer & Bègue, 2005). A recent example of research by van den Bos and Maas (2009) provides support for the dual conception model of individual BJW. van den Bos and Maas (2009) found that in an experiential mindset, negative responses towards victims were independent of a participant’s individual level of belief in a just world. Conversely, in a rational mindset only participants who strongly endorsed just world beliefs were likely to respond negatively towards the victim. These two different conceptualizations of the belief in a just world provide a new framework from which to interpret and design future research in JWT. Lastly, although strongly critical of the predominance of individual measures of the belief in a just world in the literature, individual difference measures can be used in experimental research in a supplementary manner to garner “further support for the presumed process underlying the effects of situation manipulations” (Hafer & Bègue, 2005, p. 7).
There have been several new developments in JWT that have significantly extended the scope of research in JWT beyond what Lerner had originally conceived. These include the ability to experimentally measure implicit justice responses using a modified stroop task (Hafer, 2000a) and the development of new psychometric measures (Furnham, 2003). Studies have also explored factors that moderate the relationship between the presence of an innocent victim and responses designed to reduce BJW threat proposed by JWT when manipulating, level of social self-construal (van Prooijen & van den Bos, 2009), conscious awareness of mortality (Hirschberger, 2006) and ingroup versus outgroup status (Aguiar, Vala, & Corriea, 2008). Research on JWT has also demonstrated that responses such as victim blame and derogation are not universally activated in all situations of victimization but rather are present only under certain conditions. Factors that are relevant in predicting whether blame and derogation will occur include the explicitly endorsed individual level of belief in a just world of the observer, the innocence of the victim and the persistence of the victim’s suffering (Corriea & Vala, 2003). Other factors include whether there was some kind of retribution, such as if the perpetrator of a crime was apprehended or not (Hafer, 2000a; Ijzerman & van Prooijen, 2008) and whether the victim is able to be compensated in some form for their suffering (Haynes & Olson, 2006; Kay, Jost & Young, 2005; Lerner & Simmonds, 1966).
Multiple Responses
Lerner (1980) proposed a series of nine unique strategies for coping with just world threats that fell into three distinct categories. Rational strategies accept the presence of injustice but act to prevent injustice before it occurs and restore justice to unjust situations. Rational strategies generally involve some form of positive response towards a victim of injustice, such as through helping/compensating the victim or, with foresight, acting to ensure that injustice is prevented from occurring in the first place. Non-rational strategies are characterized by a refusal to accept the reality of injustice. Non-rational strategies are best exemplified by responses towards innocent victims that serve to reinterpret the cause of the injustice through the use of strategies such as, behavioural blame or character derogation. In addition to the rational and non-rational strategies, Lerner also proposed the existence of protective strategies which are present in the form of overarching, general ways of thinking about the world such as a belief in ultimate justice where inevitably justice prevails.
Despite Lerner (1980) highlighting a total of nine different strategies for preserving a belief in a just world, the vast majority of literature on JWT has focussed primarily on non-rational, typically negative responses, particularly on the blame and derogation of innocent victims. Presumably, this is due to the direct, salient nature of both victim blaming and character derogation as well as the obvious social impact these responses can have on the well being of an individual victim and on the incidence of secondary victimization in wider society. The relatively small body of research that has looked at positive responses towards victims of injustice suggests that when given an option, individuals will respond prosocially towards victims and that, positive responses towards victims are successful in maintaining just world beliefs. For example, a study by DePalma, Madey, Tillman, and Wheeler (1999) found that in an experimental scenario involving a victim who was diagnosed with a blood disease, participants were significantly more likely to offer help in the form of their time, effort and money to a victim who was portrayed as not being responsible for the onset of their disease. They also found that high levels of helping behaviour were significantly related to the participant’s level of individual belief in a just world. Expanding on these results, Haynes and Olsen (2006) found that there was a significant interaction between the likeability of the victim, the victim’s level of behavioural responsibility for the incident and the amount of compensation offered to the victim by participants. They found that when a victim was described as having a likeable character and as having low responsibility for the incident, participants were far less likely to blame or derogate the character of the victim and instead offered large amounts of money as compensation. These studies suggest that, at least when not afforded an opportunity to blame or derogate a victim due to low behavioural responsibility or the irreproachable character of the victim, people will prefer to respond in a provictim, prosocial manner.
Alternative Strategies
Another strategy that was highlighted by Lerner (1980) that has not been fully explored in previous research is the reinterpretation of the outcome of injustice as being less harmful or perceiving the victim’s suffering in a positive light in order to reduce the threat the victim poses to just world beliefs (Hafer & Bègue, 2005). These are evident in the presence of attitudes in observers that minimize the actual suffering of the victim and stress the potential benefits to suffering such as the acquisition of important life-skills or a deeper insight into some form of meaning that ‘ordinary’ people do not have access to. This is an example of a non-rational strategy and is especially intriguing in light of the Rind and Tromovitch (1997) meta-analysis in which they argued that childhood sexual abuse was less harmful than previously thought. However, research that has included a measure of reinterpretation of suffering has failed to find any significant evidence that it is used as a just world preservation strategy (Correia, Vala & Aguiar, 2001; Kleinke & Meyer, 1990).
In summary the original assumptions of Lerner were that behavioural blame was the most preferred response towards a victim, unless the victim’s behaviour cannot be easily faulted, in which case, the character of the victim will be derogated, because it provides a more proximal or direct explanation for why the event occurred (Correia et al., 2001; Hafer & Bègue, 2005). Subsequent research findings have been mixed, with some investigators finding that blame and derogation responses are mutually exclusive (Karuza & Carey, 1984) and others finding instead that they are cumulative (Correia et al., 2001). Regardless, it is important to note that in naturalistic settings individuals’ responses are not typically uni-dimensional, there are often multiple co-occurring responses and these responses are likely the result of a complex interplay of personality traits, individual experience and unique situational variables. This may partly explain why, that in limiting the strategies available to participants to only one or two responses (e.g., blame/derogation), researchers often find hypothesized just world interactions on only some of their criteria (Hafer & Bègue, 2005). More research is needed to investigate the co-occurance and interaction effects of multiple strategies to defend just world threat beyond blame, derogation and helping.
The presence of alternative strategies for belief in a just world defence ultimately leads to questions regarding, when and for whom are different strategies preferred. This is especially important considering the potential impact that individual differences variables such as social desirability may have on participants’ responses towards victims in laboratory situations as well as their day-to-day lives. This is especially relevant when observing victims for which it is not socially acceptable to respond in a typically negative manner, (e.g., blame or derogation in the case of child victimization). Although little research has specifically addressed this question, there have been several findings that may indirectly offer an explanation. For example, some of the most robust and consistent findings in research into victim blame are the influence that demographic variables such as gender and age have on responses towards victims. For example, a study by Drout and Gaertner (1994) found significant differences in strategies used by males and females responding to a female victim of sexual assault. They found that males who scored highly on a measure of individual differences in BJW were more likely to engage in helping behaviour towards the victim, whilst females who score high on a BJW measure showed the same tendency to help but were more likely to psychologically distance themselves from the victim. Other studies exploring attributions made towards victims of sexual assault have found that male observers are significantly more likely to blame the victim than female observers (Anderson & Quinn, 2009; Foley & Pigott, 2000; Grubb & Harrower, 2009). Furthermore, other studies have found that the sex of the victim is a significant factor in predicting blame responses in adult victims (Rye, Greatrix & Enright, 2006) as well as child victims (Anderson & Quinn, 2009; Back & Lips, 1998; Hestick & Perrino, 2009) with male victims of sexual assault more likely to be held responsible for the incident than female victims. These sex differences are potentially explained by greater or lesser beliefs in rape myths, attitudes about rape that are generally false but are widely held and denying the reality of sexual aggression (Burt, 1980). Supporting this claim are findings that men also more frequently believe rape myths than do women (Grubb & Harrower, 2009; Hockett et al., 2009). Additional research has also demonstrated that in response to child victims of sexual abuse, the age of the victim is a significant factor in determining attributions of responsibility with adolescent victims being judged as being more responsible than younger prepubescent victims (Back & Lips, 1998; Davies & Rogers, 2009; Hestick & Perrino, 2009; Rogers & Davies, 2007).
Recent research has also explored specific individual differences variables that may predict the use of certain responses towards victims. One such variable investigated as a predictor for the use of the reinterpretation of suffering strategy for belief in a just world defence is a repressive emotional coping style. Despite other studies finding little support that reinterpretation of a victim’s suffering is used as a strategy for just world belief defence (e.g., Correia, Vala & Aguiar, 2002; Kleinke & Meyer, 1990) very few studies have explored the relationship between repressive coping style and the use of this particular strategy. Repressive coping is a particular personality style whereby individual’s appear immune and show little response to aversive stimuli as they are able to somehow ‘repress’ negative affect. There is evidence to suggest that repressors achieve this by conjuring up positive thoughts to repress negative responses towards threatening information (Boden & Baumeister, 1997). Supporting this are a series of studies by Hafer and Gosse (2011) who found that under conditions of low just world threat, both repressors and non-repressors were equally likely to reinterpret the suffering of the victim in a positive light, however under high just world threat repressors were still likely to reinterpret the suffering of the victim in a positive light whereas non-repressors were more likely to derogate the character of the victim. Thus it appears that repressors actively work to ‘repress’ negative stimuli that may potentially threaten their personal belief in a just world and instead interpret the outcome of a negative scenario as being positive.
Just as demographic and individual differences variables are relevant factors in determining the use of certain responses towards victims, the context in which the victimization occurred is relevant in determining attributions of blame and responsibility. Previous research into scenarios of adult rape has found that both acquaintance with the perpetrator and the presence of alcohol have significant effects on victim and perpetrator blame. For example, a victim who has been raped whilst intoxicated is attributed far more responsibility and blame than an otherwise sober victim but the inverse is true for the perpetrator who is awarded lessened blame and responsibility if intoxicated than a sober perpetrator (Richardson & Campbell, 1982; Krahe, Temkin & Bieneck, 2006). Other research looking at victim and perpetrator acquaintance have found that the greater acquaintance the victim has with the perpetrator the more the victim is blamed and the less the perpetrator is blamed (Krahe et al., 2006; Bell, Kuriloff & Lottes, 1994). The effect of acquaintance on responses towards victims is not unique to scenarios of adult sexual abuse, research into child sexual abuse has found that when comparing sexual abuse committed by a stranger or by a father, abuse committed by a father is considered more severe than abuse by a stranger. But, a father who sexually abuses his child is also seen as being less culpable and the child considered as less credible (Davies & Rogers, 2009). Little research has explored whether the type of abuse suffered by a victim is relevant in determining attributions made towards the victim and perpetrator, for example whether a case of sexual abuse is a one-off acute case or is an ongoing, chronic abuse scenario or even whether the abuse is of a physical or sexual nature. Of the few studies that have tentatively explored these questions, Muller, Caldwell and Hunter (1995) compared responses to child victims of physical abuse and adult rape victims and found that blame, derogation of the victim and abuser justification emerged as a single global blame factor. Another study found that child sexual abuse that involved penetration was considered to be far more severe and the perpetrator more culpable than sexual abuse involving genital fondling (Graham, Rogers & Davies, 2007).
Child sexual abuse
Although often overlooked, victimisation of children by adults is one interactional context that may challenge strongly held views in a just and fair world. Social attitudes have a considerable impact on a victim’s wellbeing and risk of experiencing secondary victimization, as well as legal processes and decisions, as juror attitudes and beliefs influence trial outcomes (Cromer & Goldsmith, 2010). Historically, there has been an ambivalent attitude towards childhood abuse within the medical and scientific communities, characterized by fluctuations between resistance and acceptance of the effects of child maltreatment (Dorahy, Van der Hart & Middleton, 2010). Anecdotal evidence suggests a tendency in some adults in response to childrens’ disclosures of sexual abuse to disbelieve and even express victim blame and rejection (Dorahy & Clearwater, 2012; Palmer, Brown, Rae-Grant & Loughlin, 1999). It is important to explore the social denial of child sexual abuse as this comprises an important step in providing appropriate support to victims, informing prevention and improving intervention efforts.
However the majority of research in JWT and indeed victimization in general focuses on scenarios of adult victimization. There is no plausible reason to account for the lack of interest in child victimization as theoretically the same principles of JWT which apply to adult victims may also apply to a child victim. The presence of an innocent victim constitutes a threat to the notion of the world as a just place in which people get what they deserve. Given the same threats to just world beliefs, individuals exposed to child abuse scenarios may restore justice using alternative methods or resorting to victim blame or derogation in much the same way demonstrated in adult victims. Evidence for this comes from numerous studies which have explored victim blame in the context of child sexual abuse. These studies have looked at relationships between the gender and age of both the victim and the participant (Back & Lipps, 1998; Rogers & Davies 2007), use of participants with abuse histories similar to the hypothetical victim (Harding, Zinzow, Burns & Jackson, 2010), the perceived credibility of child victims of CSA (Davies & Rogers, 2009; O'Donohue & O'Hare, 1997) and type of perpetrator coercion and victim resistance (Rogers, Davies & Cottam, 2010).
The degree to which previous research can be informed by JWT may be limited by the issue of ecological validity. Most studies utilized brief, written, hypothetical scenarios with little emotional impact for participants and can easily be seen as irrelevant to the observer’s world. Hafer and Bègue (2005) argue that a stimulus not emotionally engaging will pose little threat to peoples need to believe in a just world and will likely provoke deliberate thoughtful responses that will reflect a desire to adhere to social norms. On the other hand a stimulus that is emotionally engaging will be more likely to prime a more automatic, preconscious need to believe in a just world which will motivate attempts to restore or maintain a sense of justice. Another previously mentioned problem with current research into child sexual abuse is the frequent use of purely uni-dimensional responses towards victims such as blame/responsibility or credibility. This fails to take account for other strategies used to deal with threats to just world belief which may be preferred in different contexts (Hafer & Bègue, 2005). Another potential problem with current literature in attributions towards victims of child sexual abuse is related to social desirability. This problem clearly arises when asking participants to make strong, direct and inherently negative attributions such as blame and derogation towards children, a social demographic more typically seen as maintaining a fundamental character of  innocence. A potential solution to this problem might be to introduce more subtle, less direct measures of blame or derogation alongside more traditional, direct measures. This may allow participants greater breadth of expressing blame or derogation attributions if they are not comfortable expressing them in a more direct manner.
The Present Study
The current study explored individual’s responses towards several unique scenarios of child sexual and physical abuse. It focussed on the comparison of attributions made towards child victims of acute sexual abuse in comparison to child victims of chronic sexual abuse. A scenario detailing a case of child physical abuse was utilized as a comparison scenario to the sexual abuse scenarios. The scenarios used were designed to maximise stimulus impact and were presented in the style and format of an online news article. This was thought to heighten the potential threat that the scenarios posed to the belief in a just world and will therefore elicit stronger responses than in current CSA victim blame literature. The study used JWT as a predictive model of participant responses and allowed for the presence of multiple, co-occurring responses towards victims. The types of responses to just world threat measured were; victim blame, character derogation, psychological distancing and reinterpretation of the outcome. In addition, subtle measures of both victim blame and derogation were also included. Repressive coping style was explored as a possible predictive factor of specific responses towards the victim. The individual level of belief in a just world was assessed as a supplementary measure in order to provide additional depth
Hypotheses
This study explored whether there was a different pattern of responding towards a one-off acute case of child sexual abuse, a long-term chronic case of child sexual abuse and a case of physical child abuse. To the author’s knowledge, no previous research has explored these particular scenarios before, the direction of responding is unknown. However there are several specific predictions that can be made based on prior research.
Hypothesis One
The three different abuse-type scenarios were expected to elicit different response patterns.
Hypothesis Two
Participants are expected to be overwhelmingly pro-victim and anti-perpetrator, especially considering the age and sex demographic of the victims. It was expected that there would be relatively low levels of direct blame and derogation of the victim and high levels of blame for the perpetrator.
Hypothesis Three
Scores for the subtle measures of blame and derogation would be higher than the direct measures of blame and derogation.
Hypothesis Four
The response of psychological distancing would be high as it is a neutral response that allows participants to avoid any direct or subtle negative attributions towards the victim yet still resolves the threat the victim poses to the participant’s belief in a just world.
Hypothesis Five
Specific hypotheses relating to gender were that female participants would psychologically distance themselves from the victim more often than male participants.
Hypothesis Six
Furthermore, individuals who are defined as having a repressive coping style would be more likely to reinterpret the suffering of the victim in a positive light than individuals not defined as having a repressive coping style.
Hypothesis Seven
There is expected to be some level of interaction between the different BJW restoration strategies, it is unknown as to whether various responses will be cumulative or will be mutually exclusive.

Method
Participants
Participants were 40 undergraduate university students recruited through the University of Canterbury participant pool in exchange for course credit. In order to achieve power of 95% with a small to medium effect size of 0.375 at the significance level of α = 0.05, 36 participants were needed. In total, 41 participants took part in the study. However, one participant’s data was removed due to homogeneity of scoring on all measures. The age of participants ranged from 18 to 44 years with the majority (75%) aged between 18 to 23. 14 males and 26 females took part the in study.
Scenarios
Three different vignettes were created that detailed an act of trauma against an innocent victim, one related to physical abuse whilst the other two detailed sexual abuse. Each vignette was unique and featured a different victim and type of incident. The vignettes were written in a manner to highlight the innocence and the continued suffering of the victim in order to make sure that they pose a significant threat to observer’s belief in a just world (e.g., Corriea & Vala, 2003).  The age of the victim was deliberately left ambiguous so to avoid priming participants with information that may imply a greater or lesser level of development that may influence responding. Additionally it was important that the vignettes were presented in the format and style of a news article so that participants believed that they were genuine. This is because hypothetical scenarios that are not emotionally engaging pose little threat to the need to believe in a just world and instead elicit deliberate thoughtful responses that reflect a desire to adhere to social norms (Hafer & Bègue, 2005). The scenarios read as follows.
Scenario1:  Physical Abuse
Child hospitalized after assault by father.
A child has suffered head injuries and a broken collar bone after being assaulted by her step-father at their family home. The child was allegedly beaten with a broken broom handle, was thrown around the house "like a ragdoll" and slammed into a door with such force that it left dents in the wood. The victim is currently being treated in hospital for moderate brain injuries, a broken collar bone and extensive bruising. Doctors say that the victim may never fully overcome her injuries; she may have to face the rest of her life with, behaviour disorders, learning and social problems and possibly seizures. Members of the extended family reported that they had occasionally noticed bruising on the victim when the step-father moved in with the child’s mother around 2 years ago. Additionally, Child, Youth and Family general manager of operations John Henderson said that the family was known to CYF "through a small number of isolated incidents" but serious abuse was "never able to be substantiated".
Scenario 2:  Acute CSA
Man accused of sexual assault on young girl
A 34 year old man appeared in court today relating to charges involving a sexual assault on a local child. The man, who cannot be named because it would identify the victim, is to be charged with sexual violation of a minor, and two counts of committing an indecent act. It is understood that the accused, who was known to the victims family, lived several houses away on the same street. The incident occurred when the victim was riding her bike unsupervised on the street and was invited by the man into his house to play X-Box with him. At this point the accused is alleged to have fondled and engaged in indecent acts with the victim, telling her that her parents would get in a lot of trouble if she told anyone about what had happened. The judge said that this type of offending represented a severe breach of trust with the victim and her family and stressed the significant negative impact the offending has had on the victim’s otherwise normal development. The victim, previously an eager learner, has since become disruptive at school and aggressive towards her family at home.
Scenario 3:  Chronic CSA
Father arrested after years of sexual abuse
A 41 year old man accused of sexually abusing his step-daughter has been arrested today and is facing charges of incest, and unlawful sexual coercion. The abuse is alleged to have occurred over several years, occurring within the family home, sometimes whilst other family members were in the house. The systematic abuse started out gradually, progressing from innocent play at bedtime into fondling and then full sexual intercourse over a period of years. This situation only came to light when the victim, having difficulties at school, confessed to a school councillor details of her abuse at the hands of her step-father. It is understood that the father also attempted to restrict her social opportunities by limiting her access to friends outside of school. The consequences of the abuse have been severe for the female victim as she has developed feelings of shame and disgust at her body and become withdrawn and anxious in social situations with her peers.
Scales
Just World Beliefs
The uni-dimensional Global Belief in a Just World Scale (GBJW) developed by Lipkus (1991) was used to assess individual differences in the belief that the world is, in general, a just place where people get what they deserve. Participants indicate their level of agreement as to how well each of seven statements apply to others and to themselves. A six point likert scale is used, ranging from 1 (strong disagreement) to 6 (strong agreement). The scores for Global Belief in a Just World scale (GBJW) in this study ranged from 1 to 4.57, out of 6 with a median score of 3.35 and a mean of 3.20. The GBJW scale has a satisfactory internal consistency, α = .83 (Lipkus 1991). The GBJW scale was preferred to the more widely used Belief in a Just World Scale (JWS; Rubin & Peplau, 1975), because several studies suggest that the JWS has problems with multi-dimensionality (Hellman, 2008) and that there are different factor loadings according to gender (Hyland & Dann, 1987). In the present study the GBJW demonstrated strong internal reliability (α =  .81).

Repressive coping
The procedure used to identity participants as repressors was the same used by both Boden and Baumeister (1997) and Hafer and Gosse (2011), in which participants scoring high on the Marlowe-Crown Social Desirability Scale Short Form-C (Reynolds, 1982) and low on trait anxiety subscale of the State-Trait Anxiety Scale – Trait Version (STAI Y-2) were identified as repressors. SAY SPECIFICALLY HOW THIS WAS DONE?? Scores for the STAI Y-2 ranged from 1 to 4, with a median of 2 and a mean of 2.01. Scores for the Marlowe-Crown Social Desirability Scale Short Form-C (MC – C) ranged from 1 to 12 with a median of 6 and a mean of 5.87 with a potential score out of 13. The STAI Y-2 demonstrated high internal validity (α =  .86) and despite the MC short scale receiving adequate validity and reliability in comparison to the standard 33 item scale (see Reynolds, 1982), when used in this study it delivered a slightly questionable alpha of .65. Using the coding algorithm above to determine repressive coping, 13 participants were classified as repressors with the rest classified as non-repressors.
Dependant Measures
Blame
To measure blame/responsibility, participants were asked to indicate on a seven point scale the extent to which they believe each victim and perpetrator were to blame and were responsible for what happened. The questions used were “To what extent do you believe that the victim was at fault and to blame for the incident?” and conversely “To what extent do you feel that the perpetrator was at fault and to blame for the incident?” These questions have been used in previous studies as a measure of blame (Haynes & Olson, 2006). As the two questions measured separate constructs and were only moderately related α = .58, they were examined separately.
In addition to these two questions, participants answered a further question “Do you believe that there was anything that the victim could have done but did not do that would have stopped the incident from occurring?” This last question provides a more subtle measure that may reduce the possible effect of social desirability and allow participants an indirect means of placing at least some of the blame onto the child. This question was also analysed separately giving a total of three separate blame measures, victim blame, perpetrator blame and subtle victim blame.
Derogation
To measure derogation participants rated the victim on 8 personality traits (i.e. Considerate, Complaining, Happy, Narrow-minded, Intelligent, Insincere, Truthful and Self-centred) which have been derived from traits shown to distinguish between liked and disliked individuals (Lott, Lott, Reed, & Crowe, 1970). Ratings very similar to these have been used in previous studies to measure victim derogation (e.g., Hafer, 2000b; Haynes & Olson, 2006). In order to reduce the potential effects of response fatigue the item responses were made on a seven point scale and were structured so that half of the items had the positive pole as 7 and the other half had the negative pole as 7. In this present study the eight items used yielded acceptable levels of internal reliability (α = .75) that were similar to estimates found in previous studies; α = .78 (Hafer, 2000b) and α = .75 (Haynes & Olson, 2006).
As per the blame measure, participants also answered a subtle indirect measure of derogation as a supplement to the more direct measure mentioned above. The question participants were asked to answer was ‘How likely do you think that the event occurred, or was more likely to occur because of any particular character trait of the victim?”
Psychological distancing
Three items measured the extent to which participants differentiate themselves from the victims. Participants were asked to think back to when they were younger (or alternatively to think of somebody close to them such as a sibling or close friend) and to indicate on a 7-point scale (1 indicating highly similar and 7 indicating highly different), how similar they thought they were to the victim. The three items used were slightly reworded versions of the items used by Hafer (2000b) and had demonstrated high validity α = .95. They were “To what extent do you feel that the victim is like you?”, “In general, how similar would you say you are to the victim?” and “Overall, how much do you identify with the victim?” In the present study, these questions demonstrated high internal consistency (α = .95).
Reinterpretation of suffering
Participants’ tendencies to perceive the victim’s suffering in a positive light was measured using questionnaire items, rated on a seven point scale. The questions asked were, “How likely do you think that the victim’s experience will make them a better person?” “Are there any unexpected benefits to suffering for the victim such as the acquisition of valuable life skills?” and “For the victim is there some form of meaning or solace to their suffering?” These questions are very similar to the questions used by Hafer and Gosse (2011) to measure positive reappraisal and in the present study achieved a high cronbachs alpha of .89.
Frequency of occurrence
Participants were asked to rate how often they believed each type of incident occurs in the general population using a scale of probabilities ranging from one out of every five thousand individuals to one out of every five individuals. (e.g.  1/5000, 1/1000, 1/100, 1/50, 1/25, 1/10, 1/5).

Procedure
Participants were first asked to sign forms of consent and were given a brief overview of the nature of the experiment and what was required.  They were then given the GBJW, followed by the MC-C and the STAI Y-2  in order to identify specific participants as repressors. Participants were then presented with the three vignettes. They responded to each scenario and were afforded an opportunity to blame, derogate, distance from the victim or reinterpret the suffering of the victim. Participants were additionally asked how frequently they believed each type of scenarios occurs in the general population. In order to account for order effects, the order of presentation was randomized. Participants were then debriefed and invited to ask any questions about the nature of the study. Additionally in the debriefing, participant were asked how much they thought the vignettes were believable news articles. They were also asked which vignette they thought was the most extreme and traumatic for the victim.
Design
The design was a repeated measures MANCOVA with three levels of the independent variable (victimization: physical abuse of child by step-father, one off case of sexual abuse by neighbour, and chronic ongoing sexual abuse of child by step-father). Participant sex, emotional coping style and individual belief in a just world were used as covariates as to control for any potential interaction effects they may have on the dependant variables. The dependant measures were, the amount of blame attributed to the victim, level of derogation of the victim, level of psychological distancing between the participant and the victim and reinterpretation of the victims suffering. As a supplementary measure participants were also asked to indicate the frequency to which they believed each type of victimization occurred in the general population.
Results
Descriptive Statistics
The results between the three different conditions, comparing acute child sexual abuse, chronic child sexual abuse and child physical abuse and the dependant variables are illustrated in Table 1.
Table 1. Means and Standard Deviations of Dependant Variables in Each Condition.
	Acute Child Sex Abuse	Chronic Child Sex Abuse	Child Physical Abuse
Victim Blame	1.9 (1.5)	1.5 (1.1)	1.6 (1.2)
Subtle Victim Blame	3.9 (2)	3.8 (1.9)	2.7 (1.7)
Perpetrator Blame	6.6 (1.1)	6.8 (0.9)	6.5 (1)
Derogation	3.4 (0.8)	3.5 (0.6)	3.9 (0.6)
Subtle Derogation	4.1 (1.5)	3.2 (1.8)	3 (1.9)
Psychological Distancing	4.9 (1.6)	5.1 (1.3)	5.1 (1.1)
Reinterpretation of Suffering	2.6 (1.4)	2.5 (1.4)	2.5 (1.3 )
Rate Of Occurrence	3.1 (1.6)	2.8 (1.6)	3.8 (1.6)
Note: The scores for the dependant variables range from 1 to 7, higher scores indicate greater levels of the construct. The scores for the rate of occurrence measure ranged from one in every five thousand (1) to one in every five (7).  Standard deviations are included in the parentheses.
Overall, with the exception of perpetrator blame and psychological distancing, the means are towards the lower end of the response scale. Because the item scores were coded so that higher scores indicate greater levels of negative or anti-victim responding this would indicate that participants overall responses were strongly pro-victim. In addition, there were significant differences between the direct and subtle measures of victim blame with the subtle measures scoring higher overall than the direct measures. This indicates that there is some truth in the authors speculation that participants may prefer to respond in a more indirect manner if given the opportunity. However this only occurred between direct and subtle blame, the initial comparison between direct and subtle derogation is less consistent. The scores for the rate of occurrence measure indicate that participants believed that child physical abuse was a more common occurrence than either of the scenarios of child sexual abuse.
In addition, although secondary to the main hypotheses, some interesting correlations were found. There was a significant relationship between scores on the Marlowe-Crown Social Desirability Scale Short Form-C and the Global Belief in a Just World Scale r(38) = .35, p < 0.05. There is also an significant positive relationship between the State Trait Anxiety Scale and measures of the perceived rate of occurrence r(38) = .45, p < 0.01 indicating that participants high in anxiety were more likely to perceive that traumatic incidents occur more frequently than participants low in anxiety.
Manipulation check
As a manipulation check, during the debriefing participants were asked whether they had believed that the news articles that were presented to them were genuine or not, all participants indicated that they believed that they were genuine, although several participants (N= 3) mentioned that although they did consider it was a possibility that the vignettes were not genuine, they still believed that they were authentic because of the format of their presentation.
In addition participants were also asked to indicate which of any vignettes they had felt were more traumatic or severe for the victim. The majority of participants indicated that either the chronic sexual abuse or the physical abuse scenarios were the most severe with an approximate even split between the two conditions. This highlights a potential problem in that the acute sexual abuse vignette might not have been sufficient enough to elicit a strong enough emotional reaction or significantly threaten participants BJW.
Analyses of main hypotheses
Hypothesis One
A MANCOVA was undertaken from the data in Table 1 with type of abuse suffered (acute sexual abuse, chronic sexual abuse and physical abuse) as the independent variables, sex of participants and the individual level of belief in a just world as covariates, and the levels of victim blame, perpetrator blame, subtle victim blame, derogation, subtle derogation, psychological distancing and reinterpretation of suffering as dependent variables. The MANCOVA indicated that there was no significant change in responding between the three conditions.
However, there was a significant within subjects multivariate effect for the sex covariate F (14,134) = 1.86, p = < 0.05. Further univariate analyses indicated that this interaction is found in a significant relationship between the sex of the participant, the type of abuse the victim had received and the psychological distancing response, F (2,35) = 6.32, p = < 0.01. In addition, a non-significant trend was found between participant sex, abuse type and the derogation response F (2,35) = 3.10, p = < 0.52. These results suggest that the sex of the participant has an impact on response patterns towards a victim of abuse
Hypothesis Two
In order to further examine the participant sex effects found in the MANCOVA above, an analysis of variance (ANOVA) was used between abusetype, participant sex and the psychological distancing response. The ANOVA indicated that there was no significant gender difference in participant responding to the chronic sexual abuse condition and the physical abuse condition. However there was a significant effect of participant sex on the acute sexual abuse condition with female participants psychologically distancing themselves from the victim far more than male participants, F (2,35) = 4.12, p = < 0.05.
Hypotheses Three Four and Five
In order to determine the overall direction of participant responding and the degree to which participants utilised the different strategies for maintaining just world beliefs, a frequency table was created listing only scores that were above the neutral median (4). Comparisons of the means represented in Table 1 with the frequency scores represented in Table 2 indicate that overall direct victim blame was low and perpetrator blame was high. The scores for the subtle measure of victim blame were only moderate but were significantly higher than the scores for direct victim with responses to both of the sexual abuse conditions scoring over the median threshold. The means for the measure of derogation were also moderately high, however the frequency scores above the median were relatively low with the exception of the physical abuse condition where over half of the responses were above the median. Interestingly, the subtle measure of derogation achieved similar means and frequency scores compared to the direct measure but it was the only acute sexual abuse condition in which frequency scores were above the median. The scores for psychological distancing had very high means across all conditions and the frequency scores achieved a high representation at the high end of the response continuum with all over half the responses scoring over the median threshold in all conditions. The means for the reinterpretation of suffering measure were moderate however the frequency scores indicated that very few responses scored above the median. Lastly the scores for the perceived likelihood of occurrence measure indicated that participants believed that the two sexual abuse conditions were likely to occur to approximately one out of every hundred people but that the physical abuse condition was likely to occur to one in every fifty people in the general population.
These scores demonstrate that despite the moderately high mean scores for the measure of direct derogation, overall participant responding was pro-victim and anti-perpetrator as direct victim blame was low, perpetrator blame was high and the scores for reinterpretation of suffering was especially low. Instead participants appeared to prefer to use the subtle measures of blame and derogation as well as the neutral response of psychological distancing as a method of BJW restoration.
Table 2. Frequency Table of Scores Above the Median (4) in Each Condition.
	Acute Child Sex Abuse	Chronic Child Sex Abuse	Child Physical Abuse
Victim Blame	6	3	3
Subtle Victim Blame	24	26	15
Perpetrator Blame	38	39	38
Derogation	9	10	25
Subtle Derogation	26	17	17
Psychological Distancing	24	25	29
Reinterpretation of Suffering

Suffering	10	9	5

Hypothesis Six

The hypothesis that participants who were identified as having a repressive coping style would be more likely to reinterpret the suffering of the victim in a positive light was analysed using a MANOVA with abuse type as the independant variable, reinterpretation of suffering as the dependent variable and repressive coping status as the between subjects factor. The results of this analysis indicated that there was no significant interaction between repressive coping style and the use of reinterpretation of suffering as a means to reduce BJW threat.
Hypothesis Seven
In addition to exploring the overall pattern of participant responding through analysis of the frequency at which participant responses fell on the high end of the response spectrum a series of inter-correlations were used to determine any potential relationships between the responses available to participants. The correlations in Table 3 represent the mean of a particular BJW restoration response across each of the three abuse type conditions.
Table 3: Correlations Between the Means of the Dependant Variables Across each Condition.
	Victim Blame	Subtle Victim Blame	Perpetrator Blame	Derogation	Subtle Derogation	Psychological Distancing
Victim Blame
Subtle Victim Blame	0.45**
Perpetrator Blame	-0.32*	0.05
Derogation	0.09	-0.16	-0.13
Subtle Derogation	0.17	0.31*	0.31*	0.05
Psychological Distancing	0.17	-0.13	0.18	0.40**	-0.14
Reinterpretation of Suffering	0.43**	0.15	0.15	-0.13	0.25	0
* p < 0.05; ** p < 0.01
Analysis of these correlations demonstrates that there are many interesting interactions between different the BJW restoration strategies. However unlike previous research which found that BJW restoration strategies were cumulative (e.g Correia et al., 2001) the correlations found in the present study were neither cumulative nor mutually exclusive.  For example, the greater the level of direct victim blame, the greater the levels of subtle victim blame r(38) = .45, p < 0.01 and reinterpretation of suffering r(38) = .43, p < 0.01 however the less likely levels of perpetrator blame r(38) = -.32, p < 0.05 are likely to be. Besides the obvious negative relationship between victim blame and perpetrator blame, the total blame response including the subtle and direct measures appears to be cumulative but in addition participants may engage in reinterpretation of the victims suffering in an attempt to reduce the psychological burden of blame on themselves. Furthermore, the positive relationship between the subtle measures of derogation and blame r(38) = .31, p < 0.05 are an interesting finding considering the lack of an equivalent relationship in the direct blame and derogation measures. however the positive relationship between subtle derogation and perpetrator blame may indicate that the despite holding the perpetrator directly responsible there was some subtle factor in the character of the victim that may have contributed towards the abuse.  In addition, there was there was a strong positive correlation between the measure of direct derogation and psychological distancing r(38) = .40, p < 0.01.Thus participants who viewed the character the victim in a negative light were likely to feel that they share little in common with the victim.
Discussion
The main focus of the present research was to determine whether different patterns of responding would occur in response to different abuse type conditions of child sexual and physical abuse. Each hypothesis will be discussed in turn and additional relevant findings will also be considered. The implications of the research will also be discussed as will the limitations and the potential direction for future research.
The results of this study indicated that there was no difference in responding between the physical and sexual abuse conditions and subsequently did not support the main hypothesis. However, there was support several of the secondary hypotheses. The hypothesis that participants would be strongly pro-victim and anti-perpetrator was partially supported as although there were low levels of direct victim blame and high levels of perpetrator blame there were also moderate levels of derogation and the subtle measures of victim blame/ derogation. In addition to this hypothesis, participants were also expected to prefer to use the subtle versions of blame and derogation rather than the direct versions. Analysis of the number of participants who scored above the median (Table. 2) indicates that with the single exception of subtle derogation in the child physical abuse condition the subtle measures of blame and derogation were more popular than the direct blame/ derogation measures. Furthermore, the measure of psychological distancing was found to be especially high and was important as it was a neutral response that allowed participants to avoid any negative attributions (direct/subtle) towards the victim yet still maintained the participant’s belief in a just world. Additionally female participants also used the strategy of psychological distancing significantly more than men but only for the acute sexual abuse scenario, adding partial support to the hypothesis that female participants were more likely to distance themselves from the victim than male participants. However, in this study we found no support for the hypothesis that participants who were identified as having a repressive coping style were more likely to reinterpret the suffering of the victim in a positive light.
The analysis of participant responding indicated that there was no overall difference in response patterns between the three abuse-type conditions. This result is especially interesting as during the debriefing many participants clearly indicated that they believed that a particular scenario was more serious or traumatic than the others. However, despite this responses designed to restore the belief in a just world (e.g psychological distancing) were high across all conditions. This would indicate that although each scenario was sufficient enough to threaten participant’s belief in a just world and warrant a defensive response, differences in abuse-type did not mediate this relationship in the form specific responses related to a particular abuse type. This is in line with previous research that has found that responses towards a victim do not vary as a function of the type of abuse received (Graham, Rogers & Davies, 2007; Muller, Caldwell & Hunter, 1995).
Previous research that has explored the relationship contextual variables and responses towards an innocent victim have most often manipulated variables that are primarily relevant in determining the amount of agency a victim had to inhibit or stop the incident from occurring. For example, several studies have found that male victims are blamed significantly more than female victims in incidents of sexual assault (Rye, Greatrix & Enright, 2006) and that in the case of child sexual abuse older teenage victims are blamed significantly more than younger pre pubescent victims (Rogers & Davies, 2007). In these cases responses towards the victim may reflect participant’s social expectations of the behaviour and characteristics of the victim. For example, male victims may be blamed more than female victims because males are expected to resist and defend themselves in a manner that is not expected of a female victim in similar circumstances. Because differences in abuse-type alone do not in themselves contain any possible information that may indicate differences in the socially expected behaviour and character of the victims, responses between different abuse-type scenarios were similar.
Across the three abuse-type conditions participants were expected to be strongly pro-victim and anti-perpetrator. However the results indicated only partial support of this hypothesis. As expected there was little direct victim blame and very high perpetrator blame but there was also moderate levels of direct derogation. There were also moderate scores for the measures of subtle derogation and subtle victim blame. A potential explanation for these responses may be that due to the young age and the sex of the victim, participants believed that the victim did not have the social or physical ability to maintain any level of control over the incident. Yet they still felt that despite this in some form the victim was at least partially responsible for the incident. In response to this participants may have assigned moderately more responsibility for the incident onto the character rather than the behaviour of the victim. These responses are in line with the original assertions by Lerner and Simmons, (1966) that observer’s first looks to the behaviour of the victim for causal explanations, then if the victims behaviour is unable to be faulted then observers will assign responsibility to the character of the victim.
Furthermore, as expected other than the exception of subtle derogation in the child physical abuse condition, the subtle measures of behavioural blame and character derogation scored higher than the direct measures of blame and derogation. This finding was anticipated because of the potential influences of social desirability in making harsh negative attributions towards a child victim. The strong correlation between these two subtle measures and the previously mentioned low scores of direct blame indicate that participants are less willing to make harsh negative attributions towards a child victim but still feel that in some sense the victim is still at least partially at fault.
Building on this hypothesis, participants were also predicted to most prefer to use the neutral response of psychological distancing over all others except for perpetrator blame. The scores for this response were higher than even the subtle measures of blame and derogation.  One explanation for this is that even though the subtle responses were available and were less extreme than the direct measures they still involved participants making negative attributions that were socially undesirable to make of children and females. Thus, participants were for more likely to use a response that makes a neutral attribution towards the victim but also defends the threat to the participant’s belief in a just world. The psychological distancing response is especially interesting because of the potential impact that high levels of psychological distancing may have on other more subtle responses towards child victims of sexual abuse that are not readily  accessed by standard research tools.

Another hypothesis that in part supported was the hypothesis that female participants would psychologically distance themselves from the victim to a greater extent that would male participants. It was expected that female participants may use the psychological distancing response to a greater extent that would male participants because the victim posed a greater threat to their belief in a just world due the perceived similarity between the female participant and the female victim. However, this result is only supported in part as it was found only in the acute sexual abuse condition. The reason for this may be that on some level the acute sexual abuse scenario is slightly dissimilar to the other two scenarios. Further evidence for this lies in the non-significant but trending interaction between the sex of the participant and character derogation also only found in the acute sexual abuse condition, the trend indicated that males were more likely than females to derogate the character of the victim.
The last hypothesis that participants identified as having a repressive coping style would be more likely to use the reinterpretation of suffering response was not supported. This is a relatively new direction that research in JWT is exploring and is especially interesting as little previous research has explored specific individual difference variables that may be predictive of the use of a specific strategy for maintaining a belief in a just world. One potential explanation for this non-significant finding may lie in the poor internal consistency of the Marlowe-Crown Social Desirability Scale found in this study. Another explanation may lie in the relatively small sample size of the sample used in this study. It is possible that both of these factors contributed towards this non-significant finding.
An additional purpose of this study was to determine the inter-relationships between different responses that were speculated to be strategies for the defense of threats to an individual’s belief in a just world. Analysis of the overall pattern of responses highlighted a number of interesting interactions between different response strategies. The significant relationship between victim blame, subtle victim blame and the reinterpretation of suffering responses indicates that despite the significant differences in the usage of the two blame measures, they do appear to be cumulative. The positive relationship with the reinterpretation of suffering response might indicate that participants experience some form of psychological distress or dissonance and minimalization of the suffering of the victim may function to reduce the psychological impact of their responses. One interesting finding was the popularity of the subtle measures of blame and derogation as well as their positive relationship. The use of subtle less harsh measures alongside more commonly used direct measures are a new development in victim attribution research and may closer reflect the unsolicited responses of individuals in the general population. Another interesting finding was the relationship between psychological distancing and derogation, the high responses of psychological distancing may indicate participants unwillingness to make negative attributions onto children who might be perceived as having less control over the incident than an adult may have. However a more likely interpretation is that participants were more likely to psychologically distance themselves from an individual towards whom they have made negative characterological attributions. Additionally the finding that female participants were more likely to score higher than male participants on the measure of psychological distancing may indicate that psychological distancing precedes derogation.
Relevant to the high scores of psychological distancing was the relatively low likelihood that participants believed each scenario was likely to occur in the general population. Participants indicated that the two sexual abuse scenarios would be likely to occur to approximately one out of every hundred people in the general population but the physical abuse scenario would occur to approximately one in every fifty people. These perceived prevalence rates are incongruent with the prevalence rates found in contemporary research. For example, a recent meta-analysis found that, after reviewing 65 articles across 22 countries, 7.9% of men and 19.7% of women reportedly experienced some form of childhood sexual abuse (Pereda, Guilera, Forns & Gomez-Benitio, 2009). The low perceived rates of prevalence in conjunction with the high levels of psychological distancing found in this study indicate a motivated denial of the reality of child sexual abuse and physical abuse in the social community that functions to protect the belief in a just world of the observer.
Limitations
There were several limitations inherent in the design of this study.  The first is a limitation that is common to contemporary research in experimental psychology regarding the use of university students as a sample group which potentially reduces the generalizability of the findings. There is evidence to suggest that when compared to a general sample of blue and white collar workers, students are less likely to blame the victim of sexual harassment (De Judicibus & McCabe, 2001). In this case students may have been less likely to make negative attributions towards a child victim than a sample that was more reflective of the general population.
Another limitation of this study was that despite the manipulation between the conditions, the conditions themselves were still relatively homogeneous. For example, although the manipulation was the type of abuse received (chronic/ acute sexual abuse or physical abuse) each of the conditions still involved the innocent victimization of a child by an adult perpetrator. It is possible that the scenario of the victimization of a child was so evocative that it overshadowed participant’s responses to the variation in the type of abuse that the child received which may have caused the similarity in responding between each condition. In this sense it is difficult to explore questions such as the impact of abuse-type that is a periphery factor to the direct cause of an evocative victimization scenario. Greater difference in responding between conditions might have been achieved if the manipulation involved a factor that was central to the cause and development of the incident such as the level of innocence or culpability of the victim.
 In addition another potential limitation of this study is found in the level of measurement that was used for the response items. The items used an ordinal scale that contained a neutral category as the middle value. This may have caused overall responses to skew towards the middle of the response continuum as participant’s responses that were intended as neutral were analysed together with responses that were distinctly positive or negative.
Moreover, although the vignettes in this study were constructed to appear as if they were genuine there is a possibility that these were unable to significantly threaten participant’s belief in a just world. One potential reason why this may occur is that participants might naturally distance themselves from a stimulus that is presented in written form but may be less likely to do this when viewing a video clip or a real world interaction.
Lastly, despite the efforts to reduce this problem through randomization there is a chance that the within subjects designed allowed for a priming effect to take hold in participants. This potential effect may have been exuberated by the relative homogeneity of the abuse-type conditions. Having a larger sample size and a between subjects research design would have resolved this potential limitation.
Future Research
Mention in first paragraph that future research should explore stuff like abuse type but using methods from jwt research
Although the findings of this study did not demonstrate any significant difference in participant responding between abuse-types, there were several interesting findings in the overall pattern of responses as well as the interactions between specific responses. Future research should include dependant measures that allow for a variety of multiple co-occurring responses. Restricting participants to fewer responses ignores the wide range of unsolicited responses that naturally occur in individuals day-to-day lives. Increasing the potential range of responses towards a victim in future research will provide a more holistic understanding of the victim blame phenomenon and increases the generalizability of responses hypothesized to restore the belief in a just world.
The use of multiple responses in JWT research opens up further questions of whether particular situational variables influence the use of certain responses. The present study found that there was no change in responding towards a child victim of abuse regardless of the type of abuse the victim had received. However an explanation for these effects is that due to the extreme nature of the scenarios there was a kind of ceiling effect in that participants responded primarily in response to the interaction between the adult perpetrator and the innocent child victim rather than to the different abuse-type manipulations. Future research could potentially solve this problem by varying abuse-type over a wider variety of conditions, or through the use of more refined measures that are better suited to record small nuances in participant responding between conditions.
In addition, research in the future should expand on the use of subtle measures of blame and derogation used in conjunction with traditional measures that were used in this study. The use of subtle or indirect measures alongside the more commonly used direct measures is useful for several reasons. On the face value the psychological consequences of making an attribution that is subtle and indirect is less than the consequences of making a direct, specific and negative attribution.  This may reduce the influence of social desirability in observers who have been invited to make negative attributions of behavioural or characterological responsibility towards an innocent victim of injustice.  Another reason is that extending the range of responding to include subtle measures allows for response patterns that are more reflective of unsolicited responses by individuals in the general population. Thus the use of subtle measures in this context increases the generalizability of responses elicited from psychometric measures in JWT research. In addition despite the relatively high face validity of the subtle measures, further analysis should explore whether subtle and direct measures are a related or separate factor.













References
Aguiar, P., Vala, J., & Corriea, I. (2008). Justice in Our World and in that of Others: Belief in 	a Just world and Reactions to Victims. Social Justice Research, Vol 21(1), 50-68.
Alves. H., & Correia. I. (2008). On the Normativity of Expressing the Belief in a Just World: 	Empirical Evidence. Social Justice Research, 21, 106-118.
Alves. L., & Correia. L. (2010). The Strategic Expression of Personal Belief in a Just World. 	European Psychologist, 15(3),202-210.
Anderson. L., & Quinn. A. (2009). Gender differences in medical students’ attitudes towards 	male and female rape victims. Psychology, Health & Medicine, 14(1), 105-110.
Back, S., & Lips, H. M. (1998). Child Sexual Abuse: Victim age, Victim Gender, And 	Observer Gender as Factors Contributing to Attributions of Responsibility. Child 	abuse and Neglect, 22(12), 1239-1252.
Burt. M. R. (1980). Cultural Myths and Supports for Rape. Journal of Personality and social 	Psychology, 38(2), 217-230.
Bendig, W. (1956). The development of a short form of the Manifest Anxiety Scale. Journal 	of Counseling Psychology, 20, 384.
Boden, J., & Baumeister, R. (1997). Repressive coping: Distraction using pleasant thoughts 	and memories. Journal of Personality and social Psychology, 73, 45-62.
Braband, J., & Lerner, M. J. (1974). “A little time and effort” . . . Who deserves what from 	whom? Personality and Social Psychology Bulletin, 1, 177–179.
Correia, I., Vala, J., & Aguiar, P. (2001). The effects of belief in a just world and victim’s 	innocence on secondary victimization, judgements of justice and deservingness. 	Social Justice Research, 14, 327–342.
Cromer. L. D., & Goldsmith. R. E. (2010) Child Sexual Abuse Myths: Attitudes, Beliefs,		 and Individual Differences. Journal of Child Sexual Abuse,19(6),618-647.
Crowne, D. P., & Marlowe, D. (1964). The approval motive: Studies in evaluative 	dependence. New York: Wiley.
Dalbert. C. (2009). Belief in a Just World. In Leary, M. R., & Hoyle. R. H. (Eds.),Handbook 	of individual differences in social behavior (pp. 288-296) New York : Guilford Press
Davies. M., & Rogers. P. (2007). Attributions in a Hypothetical Child Sexual Abuse Case: 	Roles of Abuse Type, Family Response and Respondent Gender. Journal of Family	 Violence,22, 733-745.
DePalma, M. T., Madey, S. F., Tillman, T. C., & Wheeler, J. (1999). Perceived patient 		 responsibility and belief in a just world affect helping. Basic and Applied Social 	Psychology, 21, 131–137.
Dorahy, M.J., Clearwater, K. (2012) Shame and Guilt in Men Exposed to Childhood Sexual 	Abuse: A Qualitative Investigation. Journal of Child Sexual Abuse, 21, 155-175.
Dorahy, M.J., Van der Hart, O., & Middleton, W. (2010). The history of early trauma and 	abuse: From the 1850s to the current time. In R. Lanius & E. Vermetten (Eds.), The 	hidden epidemic: The impact of early life trauma on health and disease. Cambridge: 	Cambridge University Press.
Drout, C. E., & Gaertner, S. L. (1994). Gender Differences in Reactions to Female Victims. 	Social Psychology and Personality, 22, 267-278.
Ellard, J. Miller, C. Baumle, T. & Olson, J. (2002). Just World Processes in Demonizing. In 	M. Ross & D. T. Miller (Eds.), The Justice Motive in Everyday Life (pp. 350-362). 	New York: Cambridge University Press.
Epstein, S., Lipson, A., Hostein, C., & Hub, E. (1992). Irrational reactions to negative 	outcomes: Evidence for two conceptual systems. Journal of Personality and Social 	Psychology; 62, 328-339.
Foley , L. A., & Pigott, M. A. (2000). Belief in a Just World and Jury Decisions in a Civil 	Rape Trial. Journal of Applied social Psychology, 30, 935-951.
Furnham, A. (2003). Belief in a Just world: Research progress over the last decade. 	Personality and Individual differences, 34, 795-817.
Furnham, A., & Procter, E. (1989). Belief in a just world: Review and critique of the 	individual difference literature. British Journal of Social Psychology, 28, 365–384.
Graham. L., Davies. M., & Rogers. P. (2009). Perceptions of Blame and Credibility Toward 	Victims of Childhood Sexual Abuse: Differences Across Victim Age, Victim-	Perpetrator Relationship, and Respondent Gender in a Depicted Case. Journal of 	Child Sexual Abuse, 18, 78-92.
Grubb. A. R., & Harrower. J. (2009). Understanding attribution of blame in cases of rape: An 	analysis of participant gender, type of rape and perceived similarity to the victim. 	Journal of Sexual Aggression, 15(1),63-81.
Hafer, C. (2000a). Do Innocent Victims Threaten the Belief in a Just world? Evidence From a 	Modified Stroop Task. Journal of Personality and social Psychology, 79(3), 165-173.
Hafer, C. (2000b). Investment in Long-Term Goals and Commitment  to Just Means Drive 	the Need to Believe in a Just world. Personality and social Psychology Bulletin, 26,	1059-1073.
Hafer, C. L., & Bègue, L. (2005). Experimental research on Just-world Theory: Problems, 	Developments and Future Challenges. Psychological Bulletin, 131(1), 128-167.
Hafer, C. L., & Gosse, L. (2010). Preserving the belief in a just world: When and for whom 	are different strategies preferred?. In D. R. Bobocel, A. C. Kay, M. P. Zanna, & J. M. 	Olson (Eds.), The psychology of justice and legitimacy: The Ontario symposium 	(Vol. 11, pp. 79–102.). New York: Psychology Press.
Hafer. C. L., & Gosse. L. (2011) Predicting alternative strategies for preserving a belief in a 	just world: The case of repressive coping style. European Journal of Social 	Psychology,41(6), 730-739.
Harding. H. G., Zinzow. H. M., Burns. E. E., & Jackson. J. L. (2010). Attributions of 	Responsibility in a Child Sexual Abuse (CSA) Vignette Among Respondents with 	CSA Histories: The Role of Abuse Similarity to a Hypothetical Victim. Journal of 	Child Sexual Abuse, 19,171–189.
Haynes, G, A. & Olson, J, M. (2006). Coping With Threats to Just-world Beliefs: Derogate, 	Blame or Help? Journal of Applied Social Psychology, 36(3), 664-682.
Hellman, C. M., Muilenburg-Trevino, E. M., & Worley, J. A. (2008). The Belief in a Just 	World: An Examination of Reliability Estimates Across Three Measures. Journal of 	Personality Assessment, 90(4), 399-401.
Heystick. H., & Perrino, C. S. (2009). African-American Undergraduates’ Perceptions and 	Attributions of Child Sexual Abuse. Journal of Child Sexual Abuse, 18,61–77.
Hirschberger, G. (2006). Terror Management and Attributions of blame to Innocent victims: 	Reconciling Compassionate and Defensive Responses. Journal of Personality and 	Social Psychology, 91(5), 832-844.
Hockett. J. M., Saucier. D. A., Hoffman. B. H., Smith. S. j., & Craig. C. W. (2009). 	Oppression Through Acceptance? Predicting Rape Myth Acceptance and Attitudes 	Toward Rape Victims. Violence Against Women, 15(8),877-897.
Hyland, M. E., & Dann, P. L. (1987). Exploratory Factor Analysis of the Just World Scale 	using British Undergraduates. British Journal of social Psychology, 26, 73-77.
Ijzerman, H., & Van Prooijen, J. (2008) Just World and the emotional Defence of Self. Social 	Psychology, 39(2), 117-120.
Janoff-Bulman, R. (1992). Shattered Assumptions: Towards a New Psychology of Trauma. 	New York, NY: The Free Press.
Janoff-Bulman, R. (1989).  Assumptive Worlds and the Stress of Traumatic Events: 	Applications of the schema Construct. Social cognition, 7(2), 113-136.
Janoff-Bulman, R., & Yopyk, D. J. (2004). Random Outcomes and Valued commitments: 	Existential Dilemmas and the Paradox of Meaning. In Greenberg, J., Koole, S. L., & 	Pyszczynski, T. (Eds.), Handbook of Experimental Existential Psychology (pp. 122-	138). New York, NY: The Guilford Press.
Karuza, J., Jr., & Carey, T. O. (1984). Relative preference and adaptiveness of behavioural 	blame for observers of rape victims. Journal of Personality,	52, 249–260.
Kay, A. C., Jost, J. T., & Young, S. (2005). Victim-derogation and victim-enhancement as 	alternate routes to system justification. Psychological Science, 16 (3), 240-246.
Kleinke, C. L., & Meyer, C. (1990). Evaluation of rape victim by men and women with high 	and low belief in a just world. Psychology of Women Quarterly, 14, 343–353.
Lerner, M. J. (1965). Evaluation of performance as a function of performer’s reward and 	attractiveness. Journal of Personality and Social Psychology, 1, 355–360.
Lerner, M. J. (1977). The justice motive: Some hypotheses as to its origins and forms. Journal of Personality, 45(1), 1-52.
Lerner, M. J. (1980). The Belief in a Just World: A Fundamental Delusion. New York, NY: 	Plenum Press.
Lerner, M. J. (1998). The two forms of belief in a just world. In L. Montada & M. J. Lerner 	(Eds.), Responses to victimizations and belief in a just world (pp. 247–269). New 	York: Plenum Press.
Lerner, M. J. (2003). The justice motive: Where psychologists found it, how they lost it, and 	why they may not find it again. Personality and Social Psychology Review, 7, 388–	399.
Lerner, S. C. (1981). Adapting to Scarcity and Change (II) Constructive Alternatives. In 	Lerner, M. L. * Lerner, S. C (Eds.), The Justice Motive in Social Behaviour: Adapting 	to Scarcity and Change (pp. 465-472). New York, NY: Plenum Press.
Lerner, M. J., & Miller, D. T. (1978). Just world research and the attribution process: 	Looking back and ahead. Psychological Bulletin, 85, 1030–1051.
Lerner, M. J., & Simmons, C. H. (1966). Observer’s reaction to the “innocent victim”: 	Compassion or rejection? Journal of Personality and Social Psychology, 4, 203–210.
Lipkus, I. (1991). The construction and preliminary validation of a Global Belief in a Just 	World Scale and the exploratory analysis of the Multidimensional Belief in a Just 	World. Personality and Individual Differences, 12, 1171–1178.
Long, G. T., & Lerner, M. J. (1974). Deserving, the “personal contract” and altruistic  	behaviour by children. Journal of Personality and Social Psychology,29, 551–556.
Loo, R. (2002). A psychometric and cross-national examination of a belief in a just world 	scale. Journal of Applied Social Psychology, 32, 1396–1406.
Lott, A. J., Lott, B. E., Reed, T., & Crowe, T. (1970). Personality-trait descriptions of 	differentially liked persons. Journal of Personality and Social Psychology, 16, 284-	290.
O'Donohue. W., & O'Hare. E. (1997). The Credibility of Sexual Abuse Allegations: Child 	Sexual Abuse, Adult Rape, and Sexual Harassment. Journal of Psychopathology and 	Behavioral Assessment,19(4), 273-279.
Palmer, S. Brown, R. Rae-Grant, N. & Loughlin, J. (1999). Responding to Children's 	Disclosure of Familial Abuse: What Survivors Tell Us. Child Welfare: Journal of 	Policy, Practice, and Program, 78(2), 259-282.
Maes, J. (1998). Immanent justice and ultimate justice: Two ways of believing in justice. In 	L. Montada & M. J. Lerner (Eds.), Responses to victimizations and belief in a just 	world (pp. 9–40). New York: Plenum Press.
Muller, R. T., Caldwell, R. A., & Hunter, J. E. (1995). The construct Dimensionality of victim 	Blame: The situations of Physical child Abuse and Rape. Personality and Individual 	Differences, 19(1), 21-31.
Reynolds. W. M. (1982). Development of reliable and valid short forms of the Marlowe-	Crowne Social Desirability Scale. Journal of Clinical Psychology,38(1), 119-125
Rogers. P., Davies. M., & Cottam. L., J. (2010). Perpetrator coercion, victim resistance and 	respondent  gender: their impact on blame attributions in a hypothetical child sexual 	abuse case. Journal of Aggression, Conflict and Peace Research 2(3), 25-35.
Rogers. P., & Davies. M. (2007). Perceptions of Victims and Perpetrators in a Depicted Child 	Sexual Abuse Case: Gender and Age Factors. Journal of Interpersonal Violence, 	22(5), 566-584.
Rubin, Z., & Peplau, A. (1973). Belief in a just world and reactions to another’s lot: A study 	of participants in the national draft lottery. Journal of Social Issues, 29, 73–93.
Rubin, Z., & Peplau, L. A. (1975). Who believes in a just world? Journal of Social Issues, 31,	 65–89.
Rye, B., Greatrix, S., & Enright, C. (2006). The Case of the Guilty Victim: The Effects of 	Gender of Victim and Gender of Perpetrator on Attributions of Blame and 	Responsibility. Sex Roles, 54, 639-649.
Schmitt, M. J. (1998). Methodological strategies in research to validate measures of belief in 	a just world. In L. Montada & M. J. Lerner (Eds.), Responses to victimizations and 	belief in a just world (pp. 187–215).New York: Plenum Press
Spielberger. C. D., Gorsuch. R. L., Lushene. P. R., Vagg. P. R., & Jacobs A. G. (1983). 	Manual for the State-	Trait Anxiety Inventory (Form Y).Consulting Psychologists 	Press, Inc.: Palo Alto.
Van den bos. K & Maas. M. (2009) On the Psychology of the belief in a just world: 	Exploring Experimental and Rationalistic Paths to Victim Blaming. Personality and 	Social Psychology Bulletin, 35(12),1567-1578.
Van Prooijen, J. & Van Den Bos, K. (2009). We Blame Innocent Victims More than do I: 	Self-Construal Level Moderates Responses to Just-world Threats. Personality and	 	Social 	Psychology Bulletin, 35(11),1528-1539.
Weinberger, D. A., Schwartz, G. E., & Davidson, R. J. (1979). Low-anxious, high-anxious, 	and repressive coping styles: Psychometric patterns and behavioural and 	physiological responses to stress. Journal of Abnormal Psychology, 88, 369-380.