Introduction

Sexual misconduct, a term encompassing acts of sexual harassment, assault, and image-based sexual abuse (e.g., revenge porn), has been a reported issue in many, if not all, professional and public spaces (Minnotte & Legerski, 2019). Interest by the general public in understanding sexual misconduct tends to increase after allegations occur in high profile cases, such as the U.S. Senate hearings regarding Clarence Thomas’s nomination to the U.S. Supreme Court after he was accused of sexual harassment by Anita Hill (Black & Allen, 2011; Diekmann et al., 2013; Thomas, 1993), or the very recent defamation trial of Johnny Depp and Amber Heard (Garber, 2022; Winter, 2022).

The #MeToo Movement gained viral attention in 2017 after the much-publicized accusations against film producer Harvey Weinstein and the resulting attention from celebrities (Chuck, 2017). However, the term and social media tag #MeToo was first used in 2006 in an attempt to raise awareness about the large numbers of women and girls who are often the victims of sexual assault (Ohlheiser, 2017). Over time, the phrase became a hashtag shared on social media by film stars, musicians, politicians and other well-known people and the issue of sexual misconduct has become a major topic of conversation and news media focus. The #MeToo movement has been recognized as providing a space for activism and community formation and has shone a bright light on the widespread problem of sexual misconduct in professional and educational settings (Bogen et al., 2022; Kachen et al., 2020; Skewes et al., 2021). This attention has also highlighted misunderstandings and disagreements regarding definitions of sexual harassment, abuse and assault, as well as who is to blame in the events (Evans et al., 2022; Tracy & Maurer, 2019).

Although scholars have long been examining the causes and consequences of sexual misconduct in private spaces as well as the workplace (Gutek et al., 1980), prisons (Lockwood, 1979) and college campuses (Till, 1980), the public attention attracted by the #MeToo Movement has translated into a large increase in the number of scholarly and academic publications that focus on sexual misconduct. Some of this research focuses on the unique experiences of sexual misconduct based on where the misconduct occurred (Min et al., 2021), whereas other research focuses on how the movement has impacted people’s beliefs regarding disclosures of sexual misconduct (de Roos & Jones, 2022) or what people thought were appropriate punishments for the perpetrators (Nodeland & Craig, 2019).

Of particular interest, apart from but related to the #MeToo Movement, has been the scholarly interest in victim’s rights and interpersonal perceptions of the victims of sexual misconduct. Specifically, a large portion of research was prompted via an increased awareness that victims might need protection from perpetrators, with some scholarship concentrating on how the #MeToo Movement has impacted interpersonal perceptions, such as victim believability (Acquaviva et al., 2021). For example, people who were more supportive of the #MeToo Movement were less likely to see the movement as a result of hypersensitivity regarding sexual misconduct in workplaces (Smith & Ortiz, 2021), suggesting that they were less likely to view the movement as an overreaction or exaggeration by victims.

In conjunction with increased concerns about victim’s rights over the last half-century, there is also a common perception that illegitimate claims of sexual misconduct are on the rise, as well as an increase in ‘coddling’ or over-indulging victims, which some claim has led to an American culture of victimization (see Niemi and Young, 2016 for a brief review). These two opposing views (i.e., to always believe and protect alleged victims or believe that a greater number of alleged victims are making false claims) have led to controversy and disagreement. Some of this controversy has led to victim blaming, an umbrella term referring to the belief that victims are responsible for their own plights (Angelone et al., 2015; Stubbs-Richardson et al., 2018), and victim derogation, where negative characteristics or traits are attributed to the victims (Correia & Vala, 2003).

Understanding victim blaming is important because it has serious ramifications, such as upholding unequal power relations in society, allowing individuals to socially distance themselves from victims, absolves perpetrators (especially those of higher social status) from punishment, and often dehumanizes victims (Ghidina, 2019). Victim blaming can also lead to the underreporting of sexual assault and may ultimately lead to severe mental health outcomes for victims (Aborisade, 2022; Stubbs-Richardson et al., 2018). The prototypical examples of victim blaming in the literature tend to occur in cases of rape, as victims of rape are often criticized for engaging in so-called rape-inviting behaviors like previous promiscuous behavior, provocative dress, or being intoxicated (Stubbs-Richardson et al., 2018). Especially in sex related cases, relative to other crimes, victims are often met with skepticism about the events and simultaneous justification of the perpetrator’s actions (Bieneck & Krahé, 2011; Sizemore, 2013). This is particularly concerning because victim blaming likely maintains a cycle of future assault and victimization, as previous research indicates that men who engage in sexual aggression are more likely to dehumanize victims (Bevens & Loughnan, 2019), a component of victim blaming and victim derogation (Ghidina, 2019).

Even the term revenge porn, a colloquial and sometimes scholarly term used for image-based sexual abuse (i.e., the revenge-motivated recording, creation, or sharing of nude photos or sexual images without the victim’s consent), connotes victim-blaming because it implies that the victim has done something to provoke the perpetrator and to deserve revenge (Aborisade, 2022). In a recent study, participants who were victims reported that they were commonly led to believe or even told that the release of nude photos or videos without their consent was of their own doing because they were engaging in careless sexual behavior (Aborisade, 2022). The argument is that by engaging in sinful, immoral, or culturally inappropriate behaviors the victim is somehow responsible for the betrayal. As such, victim blaming is essentially a second sexual victimization via a process of dehumanization, derogation, and othering (Bevens & Loughnan, 2019; Ghidina, 2019).

In addition to victim blaming, secondary victimization can occur when an observer derogates and dehumanizes the victim by attributing negative characteristics to them, or minimizes their victimization through dismissal, social disapproval, nonconsensual disclosure, responses of shaming, intimidating, betraying, or discouraging the victim to report the situation (Correia & Vala, 2003; Jackson et al., 2017; Mendonca et al., 2016; Mulder et al., 2020). In the case of image-based sexual abuse, victims reported being subjected to condemnation, abuse from peers and family members, chastising, ostracism, and in some cases physical assault from parents (Aborisade, 2022). In another example, Mulder and colleagues (2020) found that male victims of sexual assault are often implicitly perceived as more feminine and less masculine than victims of physical but non-sexual assault. Mulder and colleagues (2020) argue that this feminization stems from a suspicion that the victim was passive and therefore implicitly allowed the victimhood.

Myriad studies have examined what attributes might lead individuals to dehumanize, blame, or derogate the victim. A recent meta-analysis demonstrated that studies using real world events (e.g., high profile cases of misconduct) reported larger victim blaming effects (Dawtry et al., 2020). Additionally, some studies have examined attributes of the misconduct, such as the length of the relationship between the victim and perpetrator (Angelone et al., 2015; Leiting & Yeater, 2017), the passivity or silence of the victim over time (Angelone et al., 2015; Diekmann et al., 2013), or the use of alcohol during the event (Leiting & Yeater, 2017), all of which can impact how and whether victims are dehumanized, blamed, or derogated.

Relatedly, other studies have examined the individual differences among participants to predict how much they engage in victim blaming, dehumanizing, and derogating behavior, such as participants’ own victimization histories and sexual attitudes (Leiting & Yeater, 2017). For example, greater adherence to traditional gender roles predicts greater levels of victim blaming and less perceived perpetrator guilt in cases of acquaintance rape (Angelone et al., 2015). This suggests that individual differences influence the way that people perceive the interpersonal behavior of others, especially those that occur within a sexual context (Hackathorn et al., 2017). As such, it is highly likely that other individual differences influence the way that people perceive those who have been accused (i.e., alleged perpetrators) as well as the accusers (i.e., alleged victims) of sexual misconduct, particularly during a time when the #MeToo Movement has focused so much public attention onto the issue (Bongiorno et al., 2020).

In the current study, we examined the influence of various individual differences, such as empathy, belief in a just world, and moral values on participants’ perceptions of the accusers, as well as the accused, in a situation of sexual misconduct. Additionally, as these interactions are by definition sexual interactions (albeit nonconsensual) we were also interested in the influence of constructs that are related to both sexual attitudes and victim derogation in past research. Specifically, we examined the predictive influence of participants’ religious identification, level of personal sex guilt, sociosexuality, and political conservatism.

Empathy

Empathy is considered a pro-social emotion directed at another person who is in need and includes other emotions such as sympathy, compassion, and tenderness (Batson et al., 2002). Part of an empathic response involves taking the perspective of the person in need and imagining how that person may be affected by the situation (Batson et al., 2002). Inducing empathic responses can help change attitudes toward an individual in need, whether or not they are responsible for their own situation (Batson et al., 2002). Moreover, research suggests that increased empathy increases readiness to help and a decrease in victim blaming and dehumanization (Batson et al., 1991; Bongiorno et al., 2020; Ghidina, 2019). In fact, myriad research suggests that empathy leads to increased value for the other person’s welfare, and provides the basis for increased motivation to help (Batson et al., 2002). Moreover, negative emotions that might be elicited through empathy for the victim predicts less victim derogation (Ash & Yoon, 2020).

However, an important caveat is that imagining one’s self in another person’s shoes does not always increase empathy or sympathy (Diekmann et al., 2013). For example, people’s propensity to imagine themselves in a sexual harassment situation, especially a situation in which the victim remained passive or silent, actually increased condemnation of the victim. This is likely because when individuals attempt to see themselves in similar situations as a victim, they often make overly optimistic errors in how they think they would behave. For example, they might imagine that they would respond actively to an inappropriate sexual comment from a coworker rather than with a more passive or silent response. Alternatively, observers who were asked to recall a similar situation in the past in which they behaved similarly (i.e., remained passive and silent) were less likely to evaluate the sexual harassment victim negatively (Diekmann et al., 2013).

In combination, these findings regarding empathy suggest that people who feel more empathy toward the accuser in a situation of sexual misconduct will likely blame and derogate the victim less than people who do not experience empathy toward the accuser. However, people who feel empathy toward someone accused of sexual misconduct (perhaps because they have personal histories that make it easier for them to empathize with the accused) may experience less empathy toward the alleged victim, thus increasing the likelihood of victim dehumanization, blaming, or derogation.

Belief in a Just World (BJW)

In a recent meta-analysis of over 50 studies across 50 years (Dawtry et al., 2020), victims are evaluated less favorably when they are a higher threat to participants’ beliefs in a just world (BJW). Just world beliefs posit that what happens to someone is related to that person’s own personal characteristics, and thus each person deserves whatever happens to them (de Judicibus and McCabe, 2001). In other words, bad things happen to bad people and good things happen to good people (Lerner & Miller, 1978). Assigning people who are the victim of catastrophic or tragic consequences with the responsibility for their own plight helps remove discomfort resulting from considering that a similar thing could also happen to the observer. As a result of the motivation to protect themselves from having to consider possible future harm to themselves from no fault of their own, people engage in cognitive transformations that deny the victim’s suffering or derogate and dehumanize the victim via character assassinations (Correia & Vala, 2003; de Judicibus and McCabe, 2001; Jensen and Gutek, 1982).

In a recent study of sexual assault victim blaming via Twitter, BJW were openly and explicitly expressed in tweets (e.g., ‘she asked for it’ or ‘he didn’t mean to’). Ideas related to BJW, such as the virgin-whore binary, were used to justify rape, reinforcing the idea that the world is orderly and fair and that perpetrators were less responsible because victims brought their victimization onto themselves in some way (Stubbs-Richardson et al., 2018). Other research suggests that the mere threat of an unjust world can increase secondary victimization, as participants attributed more negative traits to the victim when primed with an unjust world than when primed with a just world (Correia & Vala, 2003).

Moreover, individuals who are higher in BJW tend to experience greater discomfort regarding situations involving innocent victims as this is a greater threat to the just world belief system. Thus, people are more likely to derogate, dehumanize, and victim blame in such situations in an attempt to cope with that unfairness (Mendonca et al., 2016). These findings suggest that when presented with situations of sexual misconduct, participants with greater BJW will be more likely to derogate the alleged victims.

Moral Values

Over a series of studies, Niemi and Young (2016) demonstrated that moral values were the best predictor of victim stigmatization and derogation across crime types, above and beyond contributions of political ideology or religiosity. Moral values (e.g., virtues and values related to justice, harm, fairness, welfare, loyalty, and respect) influence moral judgements, which then influence judgements about victim responsibility and ascriptions of blame (Graham et al., 2011; Niemi & Young, 2016).

Moral values, as explained in the Moral Foundations Theory and measured by the Moral Foundations Questionnaire (MFQ; Graham et al., 2011), posits that there are five underlying foundations to moral judgments: caring, fairness, ingroup loyalty, authority, and purity. The first two foundations (i.e., caring and fairness) represent individualizing values, whereas the remaining three foundations represent binding values. Endorsement of individualizing moral values protect against negative attitudes toward victims because these values prohibit harm and promote impartiality. That is, individualizing values are inconsistent with the idea that some people deserve to be harmed. The more people endorse individualizing values, the more obvious their judgments are: perpetrators are to blame, not victims (Niemi & Young, 2016).

Binding values, however, focus on prohibiting group threat behaviors, such as disloyalty, disobedience, and impurities. The more someone endorses binding values, the more likely they are to treat people according to their social group memberships (such as being willing to harm outgroup members) and to be less sensitive toward victim suffering and demonstrate increased derogation. Specifically, binding values predict perceptions of victim dehumanization and a belief that victims contributed to their own situations and predict less assigned responsibility for the perpetrators (Niemi & Young, 2016). Importantly, when belief in a just world is present in conjunction with binding values, they predict greater victim blaming, especially for members of the ingroup (Albuquerque et al., 2019; Niemi & Young, 2016). Thus, it is likely that people who have greater individualizing values will be less likely to derogate alleged victims of sexual misconduct, while those with more binding values will be less likely to derogate the accused.

Religiosity

Past studies show that religion and sex have a complicated history (Hackathorn et al., 2015). However, overall trends indicate that individuals with low levels of religiosity tend to have more liberal views on sex and engage in more sexual activity, whereas individuals who are high in religiosity tend to desire less engagement in sexual behavior, especially those behaviors that include experimentation or pre-marital interactions (Hackathorn et al., 2015; Mahoney, 1980; Rosenbaum & Weathersbee, 2013).

More interestingly, religiosity is a major influence on attitudes regarding other people’s sexual interactions, especially when those interactions violate religious standards or norms. Specifically, studies have shown that one’s religiosity influence’s their perceptions of sex-related norms and attitudes, and often in an intolerant or prejudicial way (Ashdown et al., 2011; Hackathorn et al., 2022). Greater internalization of religiosity predicts more negative perceptions of victims of sex-related crimes and of sex workers (Ashdown et al., 2019; Hackathorn et al., 2022). For example, one study found that when their beliefs were salient, highly orthodox Christians tended to derogate victims; however, participants who were low in orthodox beliefs were less likely to derogate when their own Christian beliefs were salient (Lea & Hunsberger, 1990).

In a recent study, Bogen and colleagues (2022) examined over a thousand tweets, in just over five days, that specifically mentioned #ChurchToo, a religious analogy of the #MeToo movement. In many cases, the tweets not only shared who the perpetrators were, but also shared how church members condoned, ignored, or allowed the abuse to continue. Participants reported a lack of accountability by those within the church who knew about the misconduct, serious efforts to minimize the severity or deny the misconduct, and that the religious institution used or weaponized scripture or religious mores against the victims. In essence, the church itself had engaged in victim blaming and derogation (Tracy & Maurer, 2019). Taken together, these studies suggests that high religiosity may predict greater derogation of individuals who are engaging in inappropriate sexual conduct.

Sex Guilt

Importantly, there are various studies that suggest that it is perhaps the sex guilt associated with religiosity that is one of the most influential factors in negative sexual attitudes, especially in the prejudicial judgments of other’s sexual interactions (Ashdown et al., 2019; Hackathorn et al., 2017, 2022; Rosenbaum & Weathersbee, 2013). Sex guilt refers to negatively-valanced emotions related to violations of traditional standards for sexual conduct or morality (Mosher, 1968, 1979; Woo et al., 2010). Logically, sex guilt is tied to religiosity in that many religious teachings and values are concerned with appropriate sexual behavior (see Harris et al., 2008 for a review). Individuals who are high in sex guilt tend to report less sexual desire, higher negative affect related to sex and eroticism, less intimacy and engagement in premarital interactions, and not surprisingly, more conservative sexual standards (Kelley, 1985; Mosher & Cross, 1971; Rosenbaum & Weathersbee, 2013).

As sex guilt inhibits sexual behavior, it also inhibits the seeking of sexual information. Thus, there is a negative relationship between sex guilt and belief in sex-related myths, such as the character of individuals who engage in sex (Mosher, 1979). As a result, individuals who are high in sex guilt also tend to hold negative attitudes toward individuals who participate in sex-related activities, especially when those activities are perceived to break norms (e.g., Ashdown et al., 2019; Hackathorn et al., 2022). In a recent example, sex guilt was the dominant predictor in the demonization and dehumanization of sex workers, whether they were engaged in legal or illegal sex work. In fact, across two studies and three samples, sex guilt mediated the relationship between religiosity and demonization of sex workers beyond participants’ general attitudes toward pornography (Hackathorn et al., 2022).

In another example, sex guilt was a dominant predictor of the demonization of users of the website AshleyMadison.com (a website that claims to facilitate sexual affairs), even after the members of the website became victims of a massive data hack on the website (Hackathorn et al., 2017). Although, participants’ religiosity influenced their perceptions of AshleyMadison.com users, the website’s owners, and the hackers that relationship was mediated by sex guilt (Ashdown et al., 2019). In all, these findings suggest that greater levels of sex guilt may predict greater derogation because the individuals may be perceived as engaging in sexual misconduct or violating traditional sexual standards.

Sociosexuality

Sociosexuality, an individual difference trait, refers to someone’s orientation toward uncommitted sexual relationships and ranges from restricted (i.e., less comfortable) to unrestricted (Simpson & Gangestad, 1991, 1992). Past research suggests that having an unrestricted sociosexuality predicts less romantic commitment, more sexual partners, more likelihood of romantic cheating, and lower perceptions of offense related to various sexual behaviors (Hackathorn & Brantley, 2014; Mattingly et al., 2011; Simpson & Gangestad, 1991, 1992).

As it relates to sexual misconduct, Yost and Zurgbriggen (2006) found that for men an unrestricted sociosexuality was correlated with higher levels of belief in rape myths and adversarial sexual beliefs (e.g., women are sexually sly and manipulative), as well as the use of sexual aggression (Bevens & Loughnan, 2019). In other words, unrestricted males may have a higher propensity for victim blaming in sexual scenarios. For women, sociosexuality was not related to sexual aggression or belief in rape myths, but was related to dominance fantasies (Yost & Zurbriggen, 2006).

Additionally, an unrestricted sociosexuality predicts a greater likelihood of being sexually harassed as well as engaging in the sexual harassment of others. In fact, sociosexuality was a better predictor of engaging in harassment than belief in rape myths, hostile sexism, or exposure to pornography. Moreover, individuals who were unrestricted in their sociosexuality were more open to strategies of opposite-sex sexual solicitation as well as same-sex competitor derogation (Bendixen & Kennair, 2017). Simply put, unrestricted sociosexuality may relate to an increased understanding of and empathy with a harasser’s motives and, thus, greater derogation of the victim.

Conservatism

Research regarding the individual difference of conservatism has predominantly related to political values, yet shows a connection between attributions of blame and conservatism. For example, in a study examining perceptions of a theft, Williams (1984) found that conservatives blamed the victim and derogated the victim’s character more than liberals did. Across two studies, conservatives were less likely to express sympathy and more likely to express disgust directed toward the victims than were liberals. Scrutiny of victims’ obligations and responsibilities, specifically those tied to rape victims, has also been related to conservatism (Anderson et al., 1997; Lambert & Raichle, 2000; Spaccatini et al., 2019). For example, individuals with more conservative attitudes tend to also have greater rape myth acceptance (Anderson et al., 1997).

Individuals who self-reported being high in conservatism tended to blame victims more than individuals who report being more liberal (Lambert & Raichle, 2000). Additionally, individuals with a highly stringent conservative ideology, namely right-wing authoritarianism, were more likely to blame a sexualized victim of stranger harassment. Spaccatini and colleagues (2019) argued that individuals who have higher levels of right-wing authoritarianism tend to respond with intolerance and hostility toward women, especially women who defy gender roles or social norms via provocativeness. We expect that individuals who endorse conservative ideologies will be more likely to victim blame because they are more motivated to preserve tradition in society (Lambert & Raichle, 2000), and people who are victims of a crime bring to light the threats to various traditional ways of thinking.

Current Study

In the current climate, particularly regarding the #MeToo Movement and high-profile cases in the media (such as the Weinstein and Depp/Heard trials), various scholars have examined people’s perceptions of accusers of sexual misconduct. While some research has demonstrated victim blaming in the context of the #MeToo Movement (Bongiorno et al., 2020), we are not aware of much research that explores what variables influence the way that people also perceive the accused, except when specifically examining how male perpetrators and female victims are perceived differently (Bongiorno et al., 2020; see Rollero and Pagliaro, 2022 for an exception).

Previous research has examined how various constructs interact to impact how the accusers and accused of sexual harassment are perceived. For example, Milesi et al. (2020) utilized the Moral Foundations Theory to explore how moral concerns and the acceptance of sexual aggression myths interacted to predict victim blaming in cases of rape. They found that belief in the myths and moral concerns both jointly and independently predicted greater victim blaming. In related work, Rollero and Pagliaro (2022) found that greater progressivism (similar to less conservatism as we discussed above) predicted less victim blaming but more perpetrator blaming in situations of revenge porn, and that empathy mediated the relationship between progressivism and attitudes towards the victims and perpetrators.

Some previous research has demonstrated that various constructs do interact in their impact on how the accused and accusers of sexual harassment are perceived. This includes the interaction of religion and sex guilt (Hackathorn et al., 2022), moral concerns and belief in sexual aggression myths (Milesi et al., 2020), political ideology and empathy (Rollero & Pagliaro, 2022), and sociosexuality and sexual fantasies (Yost & Zurbriggen, 2006). Because our understanding of the predictors of the perceptions of accusers and accused overlap various constructs across multiple studies, our goal in this study is to bring together multiple variables from across many studies that have shown to have an impact on these perceptions and to explore how they interact amongst each other.

Based on the literature we have discussed, which theoretically and empirically connects various individual difference variables to people’s self-reported derogation toward others, in this study we investigated various research questions in an exploratory fashion. First, we explored the correlations between individual difference variables (i.e., empathy, belief in a just world, moral values, religiosity, sex guilt, sociosexual orientation, and conservatism) and the demonization of both the accusers and those accused of sexual misconduct.

Additionally, in line with recent studies which indicate that using real-world events produces greater secondary victimization effects (Dawtry et al., 2020), in the current study we used the ongoing #MeToo Movement to investigate these effects. We did this because artificial situations or fictional vignettes may not produce the same thoughtful and considerable emotional appraisal as accurately as a realistic and impactful event might; thus, we aimed to use a context that directly applies in the real world.

Method

All the research procedures and instruments were approved by the relevant IRB (BLINDED IRB #18–29) and followed ethical research practices as defined by the American Psychological Association. Data was collected in 2019 and 2020, prior to the COVID-19 pandemic, and all materials can be found on the Open Science Framework website here: <insert OSF link here>.

Participants

Students enrolled in undergraduate psychology courses at a small private college in the northeastern U.S. (N = 40) were recruited to participate in the study for partial course credit. To increase the diversity of our sample, we also recruited participants via Mturk (n = 106), a survey website maintained by Amazon. Participants who had completed 80% of the items on a survey were included in analyses that explored that particular survey, while participants who had not answered at least 80% of items of a particular survey did not have their data included in analyses for those surveys. For this reason, the number of participants whose data were analyzed from each survey varies slightly. All of the participants were residing in the United States at the time they completed the surveys.

Before analyzing our data to test our hypotheses and explore our research questions, we conducted independent samples t-tests between the two sources of our data (a college sample and a non-college, Mturk sample) to determine if there were significant differences. Because of the higher number of t-tests computed, we utilized a basic Bonferroni technique and set our criterion value for significance at p < .01 to control for Type I errors. The only variable on which the two samples were different was age (t(106.54) = 19.03, p < .001, Cohen’s d = 2.19), as the Mturk sample (M = 40.38, SD = 11.42) was expectedly older than the college sample (M = 19.05, SD = 0.78). Thus, as age was only one demographic variable of myriad, we combined the two samples into one sample for all future analyses, and age was included as a covariate in the analyses. Additionally, it should be noted that a chi-square analysis of independence between source of data and sex of the participant was not significant, χ2 = 1.09, p = .297, indicating the proportions were as expected across both samples.

In the full sample, the mean age was 34.5 years (SD = 13.6) and ranged from 18 to 69 years old. Most of the sample were white/Caucasian (76.0%, n = 111), identified as women (54.1%, n = 79), and heterosexual (84.2%, n = 123). A plurality held bachelor’s degrees (32.9%, n = 48), politically identified as Democrats (42.5%, n = 62), and were non-Catholic Christians (39.7%, n = 48). Finally, the vast majority had not been formally accused of sexual harassment (93.8%) nor of sexual assault (97.3%)Footnote 1.

Materials and Procedure

Potential participants from both sample pools were provided a link to the online survey, where they read a consent form and then, if they agreed to participate, were routed to an online questionnaire. The items in the questionnaire were identical for both sample pools; the only difference between the two groups is that the students received course credit for participating and Mturk users received US$1.00. The following measures were presented in randomized order.

Derogation of Accuser and Accused

The revised Derogation Scale (Hackathorn et al., 2017, based on van Prooijen & van der Veer, 2010) asks participants use a 7-point scale (1 = strongly disagree; 7 = strongly agree) to rate their level of agreement with five statements about people who made accusations of sexual harassment or assault (i.e., presumed victims; α = 0.92) as part of the #MeToo movement. An example item is “This situation was caused entirely by the accusers’ evilness.” The participants were then asked the same five questions about the accused (i.e., presumed alleged perpetrators; α = 0.91). For each target, the final score is an average of the ratings such that higher scores indicate greater derogation.

Empathy Toward Accuser and Accused

The Empathy Toward Targets survey (Batson et al., 1997) was modified for the purposes of this study. First, participants read the following statement: “While reading about the #MeToo movement and stories of people who claim to have been sexually harassed or assaulted, how much did you experience each of these emotions…” The participants then use a 7-point scale (1 = not at all; 7 = extremely) to indicate how much they experience each listed emotion, such as anger, happiness, and apathy. Participants completed the survey twice, once being told to indicate how much they experienced each emotion in the context of thinking about the accuser (α = 0.86) and then again in the context of thinking about the accused (α = 0.86). After reverse-coding relevant items (e.g., vengeful), items are averaged where higher scores indicate greater empathy.

Belief in a Just World

The Belief in a Just World Scale (BJW; Dalbert, 1999) has 13 statements about justice and fairness, and participants are asked to rate how much they agree with each statement on a 6-point Likert-type scale (1 = strongly disagree; 6 = strongly agree), across two subscales: General (α = 0.90) and Personal (α = 0.95). An example items in the General subscale is “I think the world is a basically just place” and a sample item from the Personal subscale is “I believe that, by and large, I deserve what happens to me.” Final scores are calculated from an average for each subscale, where higher scores on the measure indicate a stronger belief in a just world.

Moral Foundations

The Moral Foundations Questionnaire (MFQ; Graham et al., 2011) asks participants to rate how relevant certain considerations are when they determine whether something is morally right or wrong. Participants respond to 16 items (e.g., whether or not someone was cruel) using a 6-point Likert-like scale (0 = not at all relevant; 5 = extremely relevant). Then, participants respond to an additional 16 items (e.g., compassion for those who are suffering is the most crucial virtue) using a different 6-point scale (0 = strongly disagree; 5 = strongly agree). In accordance with Niemi and Young (2016), the items within the subscales of authority (e.g., showed a lack of respect for authority; α = 0.77), purity (e.g., violated standards of purity; α = 0.87), and ingroup (e.g., did something to betray his or her group; α = 0.78) are combined to create a score of binding values, where higher scores indicate more concern with prohibiting group threat behaviors, such as disloyalty, disobedience, and impurities (binding values items α = 0.91). The items within the subscales of fairness (e.g., acted unfairly; α = 0.72) and caring [harm] (e.g., was cruel; α = 0.69) are combined to create an individualizing value score, where higher scores indicate values that prohibit harm and promote impartiality (individualizing values items α = 0.82).

Religious Internalization

We revised the Christian Internalization Scale (Ryan et al., 1993) to measure a less Christian-centric religious identification by replacing phrases like “other Christians” with “other members of my faith” in order to use the survey with non-Christian as well as Christian participants. The measure has two subscales (i.e., identification and introjection), and each subscale has six items. Participants respond to each item on a 4-point Likert-like scale (1 = never true; 4 = very true). A final identification score is an average of the ratings of six items (e.g., “I pray because I enjoy it”) and higher scores indicate higher identification (α = 0.96). A final introjection score is an average of the ratings of the items (e.g., “I attend church because one is supposed to”) and higher scores represent higher introjection (α = 0.83).

Sex Guilt

The Revised Mosher Sex Guilt Scale (Janda & Bazemore, 2011) asks participants to indicate how much they agree with ten different statements (e.g., “When I have sexual dreams I try to forget them”). Participants respond on a 7-point Likert-type scale (1 = very strongly disagree; 7 = very strongly agree). After relevant items are reverse-scored and averaged, higher scores on the measure indicates greater levels of sex guilt (α = 0.90).

Socio-Sexuality

The Socio-Sexual Orientation Inventory (SOI; Simpson and Gangestad, 1991) consists of a mix of seven open-ended (e.g., “With how many different partners have you had sex (sexual intercourse) within the past year?”) and various Likert-type items [e.g., “Sex without love is OK” answered on a 1 (strongly disagree) to 9 (strongly agree) scale]. Due to the difference in item types, participants’ open responses are recoded into an ordinal scale that helps in maintaining normality (Penke, 2011). The items are then summed according to instructions provided by Simpson and Gangestad (1991, and higher scores indicate more comfort with sex outside of committed relationships (α = 0.80).

Conservatism

The Core Conservatism Scale (CCS; Solomon & Harvey, 2011) asks participants to indicate their level of agreement with 12 statements regarding conservative political ideologies on a 7-point scale (1 = disagree very strongly; 7 = agree very strongly). The CCS has three subscales that measure, respectively, different aspects of conservatism based on individualism, attitudes toward social change, and attitudes toward equality. Final scores for the subscales as well as the entire survey are averaged, where higher scores indicate greater conservatism (α = 0.92).

Social Desirability Responding

To control for socially desirable responding by participants, we utilized the impression management subscale of the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1984). This subscale has 20 items on a 7-point Likert-like scale (1 = not true; 7 = very true). These items, such as “I sometimes tell lies if I have to” and “I never swear” measure participants’ likelihood to respond in a way influenced by social norms to create a positive impression of themselves rather than provide accurate responses. After reverse-coding relevant items, items are recoded such that extreme responses of 6 or 7 are awarded one point each. Then participants’ final scores are summed, which can range from 0 to 20 (M = 5.89, SD = 4.35). Extreme outliers (three standard deviations or farther from them mean) are usually removed from the data; however, in this case there were no extreme outliers (α = 0.85), and scores on this scale were not related to the dependent variables in this study (ps > 0.05).

Demographics

The demographic questions included open-ended questions about age, gender, sexual orientation, whether the participant had been accused of sexual assault/harassment, if they had accused someone of sexually assault or harassment, and ethnicity/race.

Qualitative Perceptions of #MeToo

Finally, participants were asked to write approximately 200 words about their thoughts regarding the #MeToo movement. Specifically, they were asked to “please tell us what you think about the #MeToo movement.”

The majority of participants completed the questionnaire in less than 30 min. Upon completion, participants received a brief post-research educational script describing the research and expected outcomes.

Results

To test our expectations that many of the variables in our study would be interrelated, we first computed a series of Pearson’s correlations, bootstrapped at 1,000 samples. As can be seen in Table 1, each significant correlation (at p < .05 level) is marked with an asterisk (*) and correlations that are significant at a p < .01 level are bolded. To control for the possibility of committing Type I errors, we applied a Bonferroni-like approach and only focused our follow-up analyses on factors that were statistically correlated at a p < .01 criterion with our main DVs of interest (i.e., derogation toward the accusers and accused of harassment/assault); however, the full correlation matrix can be seen in Table 1.

Table 1 Correlation coefficients for all factors
Full size table

As can be seen in Table 1, higher scores on empathy toward either the accused or the accuser is inversely correlated. That is, higher empathy toward the victim and lower empathy toward the accused is related to less victim blaming or victim derogation. The opposite relationship exists in regards to demonizing the accused. That is, higher empathy toward the accused and lower empathy toward the accuser was related to less demonization of the alleged harasser. Additionally, victim demonization was positively related to religious introjection, conservatism, and binding values, whereas demonization of the accused was related to more individualizing values and restricted sociosexuality.

As a follow-up exploration to examine which of the significant correlates might be more robust predictors, we conducted two multiple regressions that regressed each of the main dependent variables (i.e., derogation toward the accuser and derogation toward the accused) onto the significant correlates. The multiple regression predicting greater derogation of the accuser indicated that the largest predictors were less empathy toward the accuser, greater empathy toward the accused, and religious introjection, F(9, 105) = 3.40, MSE = 2.17, p < .001, R2 = 0.23. See Table  2for the coefficients. Additionally, the multiple regression predicting greater derogation of the accused indicated that the largest predictors were greater empathy toward accuser and less empathy toward the accused, F(9, 106) = 5.23, MSE = 1.72, p < .001, R2 = 0.31. See Table 3 for the coefficients.

Table 2 Regression coefficients predicting derogation of the accuser
Full size table
Table 3 Regression coefficients predicting derogation of the accused
Full size table

Qualitative Data Coding and Analysis

As part of our study, we also included one exploratory open-ended question for our participants, asking them, in less than 200 words, to “please tell us what you think about the #MeToo movement.” Of the 146 participants who completed the quantitative measures, 137 of them provided an answer to this question. To analyze this data, we followed the suggestions laid out for thematic analysis by Braun and Clarke (2006). In this process, three members of the research team read through all the participants’ qualitative responses to familiarize themselves with the data. They each independently developed initial codes, focusing on patterns, themes, and attention-catching aspects of the responses. The three coding researchers then came back together and worked to collectively develop larger, overlapping themes based on their individually-created codes. These themes encapsulated the codes that the three researchers had in common, combining codes into themes based on similarity. Codes that were not common across the three researchers or that were only seldom present in the data were eliminated.

This process created four umbrella themes, and each umbrella theme has a small number of more specific subthemes. After the themes had been defined, each of the three researchers again independently coded all the data from all of the participants. The data was coded with a “1” if it contained a particular umbrella theme and a “0” if it did not. If a response contained an umbrella theme (that is, was coded as a 1 for the presence in that participant’s response of that umbrella theme), then at least one of that umbrella theme’s subthemes had to also be coded as present. In this way, each participant’s response was coded for the presence or absence of each umbrella theme and each subtheme. Once the three researchers had completed this independent coding, Cohen’s kappa was computed to ensure interrater reliability. Only codes that had a kappa of 0.85 or higher were considered reliable. For unreliable codes, the researchers met to discuss and redefine the codes and then independently coded the data again. This process was continued until the coding for each umbrella theme and subtheme achieved a minimum kappa of 0.85.

See Table 4 for the proportion of comments that were coded for the presence of each umbrella theme. Responses that were coded for the presence of an umbrella theme were also each coded for the presence of at least one of that umbrella theme’s subthemes. Table 4 contains the proportions, definitions of umbrella and subthemes, and a response from a participant that exemplifies that particular theme.

Table 4 Regression coefficients predicting derogation of the accused
Full size table

We computed various χ2 to explore how the themes in the participants’ responses interacted with our variables of interest. To do this, we first created categorical variables of high derogation, low derogation, high empathy, and low empathy for both the accuser and accused by using the median score of each of these variables to split the participants based on their scores on the relevant surveys. Thus, participants were placed into various categories based on their derogation and empathy scores, meaning each participant was assigned to one of two categories regarding derogation of the accused (high derogation or low derogation), one of two categories regarding derogation of the accuser (high derogation or low derogation), one of two categories regarding empathy toward the accused (high empathy or low empathy), and one of two categories regarding empathy toward the accuser (high empathy or low empathy).

Median scores were calculated for both the demonization of the accuser (Med = 2.00) and the accused (Med = 4.00) as well as empathy toward the accused (Med = 2.95) and accuser (Med = 4.00). Creating categories in this way allowed us to explore if the categories of derogation and empathy to which participants were assigned were independent of the ways in which their qualitative responses were coded.

We conducted four χ2 of independence for each of the four umbrella codes to determine if the participants’ qualitative responses were independent from the derogation and empathy categories they were assigned. That is, we conducted a χ2 between Umbrella Code One (i.e., whether or not that code was present in the data) and derogation of the accuser (high versus low). We then computed three more χ2 for Umbrella Theme One: (1) derogation of the accused; (2) empathy for the accuser; and (3) empathy for the accused. This process was repeated for each of the four umbrella themes.

Because the subthemes are included in and not independent from the umbrella themes to which they belong, we only conducted χ2 analyses on the subthemes if the χ2 for the umbrella theme was significant. For example, if the χ2 for Umbrella Theme One and derogation toward the accuser was not significant, we did not compute any χ2 analyses for the subthemes of Umbrella Theme One for that derogation category. Finally, because of the high number of χ2 analyses we computed, we used a Bonferroni-like technique to reduce the chances of committing a Type I error, and only accepted significant tests at the p < .01 criterion.

None of the four χ2 analyses for Umbrella Theme One were significant, and so we did not compute χ2 analyses for any of that umbrella code’s subthemes. None of the χ2 for Umbrella Theme Three were significant, nor were any of the χ2 for Umbrella Theme Four (that is, at the p < .01 level; there were some tests below the p < .05 level). Because the χ2 for the umbrella codes were non-significant, we did not compute any follow-up χ2 analyses for any of the subthemes related to Umbrella Themes One, Three, or Four, and we do not include further details on these non-significant results.

However, the χ2 of independence between high versus low empathy toward the accuser and Umbrella Theme Two was significant (χ2 = 10.55, p < .001), as was the χ2 between Umbrella Theme Two and high versus low empathy toward the accused (χ2 = 15.79, p < .001). The χ2 analyses between high and low derogation for both the accuser and the accused and Umbrella Theme Two were not significant. We followed up the significant χ2 analyses between Umbrella Code Two and the empathy scores by computing a χ2 for each of the subthemes of Umbrella Theme Two and the empathy scores. Because only Umbrella Theme Two has any significant relationship with levels of empathy or derogation, only Umbrella Theme Two is presented in Table 5.

Table 5 Regression coefficients predicting derogation of the accused
Full size table

As can be seen in Table 5, the χ2 of independence for subtheme 2.1 (“empowerment”) and the level of empathy toward the accused is significant, and it appears that those participants who were categorized as high empathy toward the accused were much less likely to include issues related to empowerment in their qualitative responses. In addition, the χ2 exploring the independence between empathy toward the accused and subtheme 2.3 (“about time”) was also significant. It appears that people who were categorized as high empathy toward the accused were much less likely to include in their qualitative responses’ issues related to how it was ‘about time’ to have the #MeToo Movement.

Discussion

As we mentioned in the literature review, our main goal in this study was to bring together the various constructs that previous research has indicated impact the perceptions of the accusers and accused of sexual harassment. While various studies had investigated the predictive ability of one or two variables on these perceptions, this is the first study we are aware of that attempted to bring together so many of these variables into one predictive model in order to determine their individual impacts as well as how they interacted with one another. In order to explore the impacts these variables have, we discuss our findings in two main ways: how these variables interact with each other and with empathy to predict perceptions of the accusers (alleged victims) of sexual harassment, and how they interact with empathy to predict perceptions of the accused.

First, though, our results indicate some interesting factors that may correlate with the behavior of dehumanizing and derogating a target, whether the victim or the perpetrator. The likelihood of demonizing the accuser (i.e., the alleged victim) was correlated with higher empathy for the accused (i.e., alleged perpetrator), lower empathy for the accuser, higher conservatism, higher moral value placed on ingroup and purity, and religious introjection. The likelihood of demonizing the accused was correlated with higher empathy for the accused, lower empathy for the accuser, higher moral value placed on fairness and avoiding harm, and a restricted sociosexuality. Contrary to past research and expectations, the likelihood to dehumanize and demonize either party was not related to beliefs in a just world, internalized religiosity, or gender of the participant.

A series of multiple regressions indicated that, overall, our results clearly demonstrate that empathy toward both the accuser and the accused in a situation of sexual misconduct plays a strong role in whether and how much participants derogated the accuser and/or the accused in the situation. Our findings support past research that empathy is correlated with a decrease in victim blaming (Batson et al., 1991; Bongiorno et al., 2020; Ghidina, 2019), but we also further the literature by showing the inverse relationship that exists when an individual has empathy for the other actor in the situation. This is an important distinction to include in future victim-blaming and dehumanizing research. For example, future research could explore the question of which is the more important predictor – empathy for the victim, or a lack of empathy for the perpetrator?

Empathy and Derogation of the Accuser

In order to understand and potentially increase empathy (and decrease derogation and dehumanization) toward people who make accusations of sexual misconduct, further attention could be paid to the variables that our data demonstrate have relationships with these two constructs. For example, political ideology and the individualizing moral values (i.e., avoiding harm/care and fairness) were correlated with greater empathy toward the accusers and a greater likelihood to demonize the accuser. Participants who were less conservative and had lower individualism scores tended to have greater empathy toward the accusers. This relationship between conservatism and empathy has been documented in previous literature in various contexts (for example, see Feldman et al., 2020; Jami et al., 2019; Morris, 2020; Sparkman et al., 2019). Although it is important to remember that our results are correlational and not causal, they do raise interesting questions that future researchers could explore about how pre-existing political beliefs, and perhaps the way politicians and pundits discuss issues of sexual misconduct, might influence people’s empathy toward and ultimately their derogation of alleged victims of sexual misconduct.

Previous research has suggested that moral values are one of the strongest predictors of victim blaming, derogation, and dehumanization (Nieme & Young, 2016). This is true regardless of the type of crime or assault, and moral values are a stronger predictor than religiosity or political ideology. Our findings show that participants who had more individualized values (e.g., placed higher value on care, reducing harm, and fairness) had greater empathy toward the accuser, although these values were not correlated with derogating the accuser. Instead, binding moral values (e.g., placing value on ingroup membership and purity) were correlated with the likelihood of demonizing the accuser, but unrelated to any empathy.

Moral foundations based on fairness, purity, and avoiding harm were all positively correlated with greater derogation of people accused of sexual misconduct. This suggests a strong relationship among fairness, avoiding harm, and perceptions of people involved in situations of sexual misconduct. Higher levels of fairness and moral harm avoidance both positively correlate with empathy toward the accuser (as discussed previously) and with derogating the perpetuators of sexual misconduct. Thus, people who utilize fairness and avoiding harm as important aspects of moral thinking decision making seem to have strong beliefs about supporting the accusers and derogating the accused. Examining causal links among these relationships is a ripe field for future research.

Finally, this suggests that the source of one’s moral values might have an indirect relationship with both empathy and derogation (relationships that might themselves be mediated or moderated by variables we did not include in our study). Future research could more closely examine the relationships among which values are important in people’s moral decisions – a line of research that might suggest potential pathways to increasing empathy via interventions.

Empathy and Derogation of Accused

It may seem counterintuitive to be concerned with empathy toward alleged perpetuators of sexual misconduct in the same article in which we discuss empathy toward the victims of sexual misconduct; however, if one has a goal to decrease instances of sexual misconduct, then it is vital to understand not only why perpetuators engage in sexual misconduct but also why others might have empathy toward them. As the qualitative data we collected shows, more than 25% of participants’ open-ended responses showed some type of negativity or hostility toward the victims in the #MeToo movement or toward the existence of the movement in general. For example, one participant wrote, “The referenced movement is an absolute joke and only exists to get special favors and treatment, while ignoring the fact that men and women each have different roles to play in society.” Other quotes listed in Table  4 demonstrate how various participants claimed the movement was hurting innocent people, that accusers were making accusations for personal gain or fabricating accusations, and that accusers are simply part of ‘victim culture.’ Understanding why some participants might take these views is important because those views could be related to a lower likelihood of holding perpetuators accountable and supporting programs that decrease misconduct (for example, see Smith and Ortiz, 2021).

Participants who had less empathy toward the accused were more likely to write about issues related to victim empowerment in their open-ended responses – but people who had more empathy toward the accusers did not mention empowerment as often. This suggests an interesting finding related to the specific target of the participants’ empathy rather than simply higher or lower general levels of empathy. It was participants who had lower empathy toward the accused perpetrator (rather than greater empathy toward the alleged victims) who talked more about empowerment. Future research should explore how the target of empathy might impact specific reactions to victims and perpetrators of sexual misconduct, as our findings show that less empathy toward the accused did not necessarily lead to the same outcomes as greater empathy toward the accuser.

Finally, we did ask participants if they had ever formally accused or been formally accused of sexual assault and/or harassment. Only seven (4%) of the participants answered affirmatively (six claimed they had formally accused someone, and one said they had been formally accused). Because this is such a small number of participants, comparing their responses to the other 96% of participants would be statistically inappropriate. However, we decided to leave these seven participants in the dataset in order to make sure we were including diverse perspectives regarding people’s personal experiences with sexual harassment and/or assault (see Footnote 1 for more details about this).

It would be interesting to conduct a similar study that has a larger proportion of those who have formally accused someone or been accused (or, perhaps, a sample solely comprised of such participants) to explore if and how that experience might impact their perceptions of accusers and accused. It would also be helpful to ask such participants if those accusations made against them were accurate or not (though people accused of misconduct typically deny the veracity of the accusations while fewer than 10% of accusations are shown to be false; Whiting, 2019) to determine if people who maintain their innocence would be more likely to derogate accusers.

Whether participants who were accused of misconduct were falsely or accurately accused, however, the experience of being accused is not pleasant and can have reputational, economic, relational, and career-related consequences to individuals and institutions (Besley et al., 2021; Bouzzine & Lueg, 2021). Because of this, it makes sense that participants who had that personal experience would experience greater empathy towards others they perceive may be having the same experience (Alexander et al., 2020) while perhaps derogating or dehumanizing those they feel are responsible for creating that experience. Future research could explore the impact that previously-accused people have in the public conversation about sexual misconduct to examine if they have an influence on the way accusers and accused are not only represented in the media (both positively and negatively) but also in the public’s collective mind.

Relatedly, sexual harassment and assault are notoriously underreported (Scurich, 2020). We asked our participants if they had ever formally accused someone of sexual harassment or sexual assault (or been formally accused). This likely means that there were participants in the study who had been harassed or assaulted but not reported it, as well as participants who had been the perpetrators of harassment and assault but never reported. Future research should consider how to handle this issue of underreporting. In addition, future research could explore the impact of participants knowing someone who was either a victim of or someone accused of sexual harassment and/or assault on their perceptions of accusers and the accused (see Sorenson et al., 2014).

Finally, previous research shows that gender plays an important role in victim blaming, derogation, and dehumanization. For example, Bongiorno et al. (2020) reported that participants who took a male perpetrator’s perspective rather than a female victim’s perspective were more likely to have greater empathy for the perpetrator than the victim. Other research (Ash & Yoon, 2020) suggests that women are generally less likely to derogate or dehumanize victims than are men. In our study, however, the gender of the participant was not a significant covariate. This is actually in line with Bongiorno and colleagues’ work, who found that while the gender of the person whose perspective participants took did impact participants’ empathy toward a victim in a vignette about sexual misconduct, the gender of the participants themselves did not have an impact. Future research should continue to examine how gender – and, more specifically, the acceptance of social and cultural gender roles – impacts the way that participants blame, derogate, and dehumanize victims of sexual misconduct.

Overall, our results are complicated and intricate. These findings suggest the likelihood of complex relationships among many different constructs when it comes to understanding empathy and derogation of both alleged victims and perpetuators of sexual misconduct. For example, perhaps some of the variables we measured were correlated with empathy or derogation only indirectly (e.g., that relationship itself is mediated or moderated by another variable we did not examine), and so the relationship does not remain significant in the regression models we tested. Future research should continue to explore the complicated web of relationships among these variables via larger and more inclusive path analyses. Doing so will further clarify both direct and indirect relationships among relevant variables, better disentangle the complex relationships among the constructs, and possibly allow for interventions or programs that increase empathy and thus decrease derogation and dehumanization, particularly of alleged victims of sexual misconduct.

Conclusion

Our findings demonstrate a clear link between empathy and derogation of both the accusers and accused in situations of sexual misconduct. In general, people who feel more empathy toward one or the other person involved in such a situation are also less likely to derogate or dehumanize that person. Greater empathy toward one person in the situation is also directly and significantly correlated with greater derogation toward the other person. While other variables that were correlated with empathy and derogation were not significant aspects of the regression models we tested, the pattern of correlations, particularly regarding moral foundations, suggests that more direct research into how these constructs impact empathy and derogation is warranted. Finally, the qualitative data from participants’ open-ended responses reinforce the quantitative data – particularly on issues related to the connection between empowerment and empathy – and also underscore the relatively large number of people with negative and hostile perceptions of the #MeToo movement.

The findings and conclusions from our study can be used not only to spur further research into the predictors and correlates of empathy and derogation of accusers and accused, but can shed light on possible interventions or programs to increase empathy and support for alleged victims of sexual misconduct and to understand why people might empathize with alleged perpetuators in order to decrease rates of sexual misconduct. As policy makers, researchers, human resources leaders, and educators better understand the complex relationships among moral foundations, past experiences, political ideology, empathy, dehumanization and derogation, we will be more successful at decreasing instances of sexual misconduct and responding appropriately and effectively when they do occur.