Ethnic Differences in Support for Restorative Justice in Sexual Abuse Cases: The Role of Offense Severity and Malleability Beliefs

Abstract

This study explores public attitudes toward restorative justice (RJ) in cases of intrafamilial and extrafamilial sexual abuse within the Israeli multiethnic society. The research investigates the role of perceived offense severity and malleability beliefs, moderated by ethnic identity. Findings reveal that Jewish participants perceive intrafamilial offenses as more severe and show lower support for RJ, mediated by malleability beliefs. In contrast, Arab participants do not differentiate between intra- and extrafamilial offenses in terms of severity, and their support for RJ is less influenced by these factors. These results highlight cultural factors shaping support for RJ in sexual abuse cases.

Introduction

A substantial body of research has explored public perceptions of individuals who have committed sexual offenses (e.g., Harper & Bartels, 2017; Hartley & Bartels, 2022). These investigations include diverse groups, such as the general populace (McCoy & Gray, 2007), students (Fuselier et al., 2002), and law enforcement officers (Kite & Tyson, 2004), and cover multiple countries (e.g., Bartels et al., 2021; Sparks, 2021; Wurtele, 2021). The evidence suggests that the public generally holds very negative attitudes toward those convicted of sexual offenses (Richards & McCartan, 2018; Weimann-Saks et al., 2022). Society perceives even “nonviolent” individuals who have committed sexual offenses as posing “a far more immediate menace than the mugger, robber, murderer, confidence trickster, or corporate polluter” (Jenkins, 1998, p. 200). In fact, individuals who have committed sexual offenses face hatred “unmatched by attitudes to any other offenders” (Sampson, 1994, p. 124). Consequently, the public is more inclined to support restricting their civil liberties than those of any other type of offender, including armed robbers and murderers (Pickett et al., 2013). These negative attitudes significantly impact offenders, fostering internalized stigma, which can lead to diminished self-worth and reluctance to seek rehabilitation (Moore et al., 2018). They also influence legislation and policies by reinforcing punitive approaches that limit opportunities for reintegration (Clark, 2007; Kernsmith et al., 2009; Mancini et al., 2010).

Recent studies have begun to investigate the impact of these harsh attitudes on public perceptions regarding the implementation of restorative justice (RJ) processes in sexual offenses (Marsh & Wager, 2015; Peleg-Koriat & Klar-Chalamish, 2020). RJ is a way of doing justice following an offense that is primarily oriented toward repairing individual, relational, and social harm (Walgrave, 2013). Redressing the harm caused by the offense may be achieved by unprejudiced, guided dialogue between the parties affected by the offense on what the participants need to do to address the needs of those affected by the offense (Zehr & Mika, 2003).

This research contributes to this literature by examining attitudes toward RJ in cases involving adolescent victims of sexual abuse. Specifically, it focuses on scenarios where a female adolescent victim reports an assault perpetrated by an adult male after she has reached adulthood. The current study further distinguishes between offenses committed within and outside the family. Finally, for the first time, it explores the influence of participants’ ethnicity, their perception of the offense's severity, and their malleability beliefs on the implementation of RJ processes.

Restorative Justice in Sexual Abuse Cases

RJ typically involves a meeting between the victim, the offender, and, if appropriate, family members, friends, and community representatives, with a primary focus on addressing the needs of all participants. During the process, the offender must confront the consequences of their actions, recognize the harm they have caused, and explore ways to make amends (Roche, 2003). Effective harm alleviation occurs when the offender adopts a reintegrative approach by acknowledging their wrongdoing, understanding the severe hardships experienced by the victim and others, and showing willingness to address the resulting needs (Zehr, 2002).

Braithwaite's (1989) reintegrative shaming theory, which is central to the RJ approach (Harris & Maruna, 2005), distinguishes between reintegrative and disintegrative shaming (stigmatization). The former is based on a process that condemns the act while accepting the perpetrators and showing concern for their rehabilitation. According to this approach, shaming during the RJ process must be reintegrative and not stigmatizing, as in legal procedures that often prevent convicts from reintegrating into the community. Reintegrative shaming uncouples self and behavior so that the perpetrator's self is sustained “as sacred rather than profane” (Braithwaite & Mugford, 1994, p. 146).

To facilitate reintegration and ensure the process's success, RJ meetings require preparation, typically ranging from 1 month to 2 years. Facilitators ensure a safe environment by considering the victim's psychological readiness and the offender's sincerity and remorse (Klar-Chalamish & Peleg-Koriat, 2021). These preparatory meetings assess safety needs and the potential scope of dialogue (Jülich, 2006; Koss, 2014). Only then do the parties engage in a structured, facilitated discussion about the offense and its impact (Umbreit & Armour, 2010). In the process, offenders often experience a significant shift in recognizing their responsibility and making reparations (Gal, 2015). A successful process entails redressing the harm caused and turning over a new leaf for all involved (Hayes & Daly, 2003).

The specific challenges of sexual offenses, including their underreporting and the inadequacies of the criminal justice system, have led to the belief that RJ processes can provide more effective outcomes. These processes foster healing and growth by addressing survivors’ needs, offering a less intimidating alternative to the formal legal system, and promoting offender accountability (Jülich, 2010; Koss, 2014; Koss & Achilles, 2008). RJ allows victims to share their stories in a safe and supportive environment (Daly & Curtis-Fawley, 2006; Jülich, 2006; Marsh & Wager, 2015). Victims receive acknowledgment and validation of the harm, public condemnation of the offense, and relief from guilt and shame (Herman, 2005), in a respectful manner (Stern, 2010). Moreover, the process empowers victims by allowing them to exercise control (Daly & Curtis-Fawley, 2006; Jülich, 2010; McGlynn et al., 2012). Unlike legal procedures, RJ depends on the offender's assumption of responsibility from the outset. Participation in RJ, where the offender admits guilt and the victim is supported by loved ones, facilitators, or victim advocacy groups, can reduce self-blame and aid in recovery and healing (Daly & Curtis-Fawley, 2006; Miller, 2011; Toren, 2015).

Moreover, the RJ process focuses on the offense and its restoration. Rather than adhering to strict rules, it is tailored to the needs of participants, emphasizing the victim's voice, choice, and control—elements that empower victims and contribute to their recovery (Braithwaite, 2002). In a study by Klar-Chalamish and Peleg-Koriat (2021), participants often described RJ as a turning point in their healing, providing a platform for agreements regarding future interactions within their family and close social environment in a constructive, forward-looking manner. This aligns with studies demonstrating RJ's role as an alternative path to justice that fosters recovery, growth, and development (e.g., Daly & Curtis-Fawley, 2006; Jülich, 2006, 2010; Koss, 2014; Koss & Achilles, 2008).

Despite the potential benefits of RJ in cases of sexual offenses, there are some concerns. RJ generally requires the offender to acknowledge their accountability and the consent of both victim and offender—two necessary conditions. Therefore, RJ cannot be a universal alternative to a formal legal system (Mika et al., 2002; Shapland, 2014). However, the development of RJ for sexual abuse aims not to replace the criminal justice system but to provide an alternative or complementary process to a formal legal system.

Another concern is the power imbalance between victim and offender inherent in sexual abuse, complicating the pursuit of justice (Daly & Curtis-Fawley, 2006; McGlynn et al., 2012). RJ for sexual abuse seeks to achieve justice for both victims and offenders, requiring facilitators to intervene during meetings to address offenders’ explanations regarding the abuse based on gender differences, victim-blaming, or denial (Jülich, 2010). RJ aims to balance power through fair processes, support for the “weaker” party, and challenges to the “stronger” one (Morris, 2002).

Public Attitudes Toward Restorative Justice in Sexual Abuse Cases

RJ processes in cases of sexual offenses also face a significant hurdle in the form of public attitudes. In recent years, public attitudes toward crime, punishment, and alternatives to punishment have become a significant focus within criminological research (Kutateladze & Crossman, 2009; Payne et al., 2010). Policymakers frequently shape crime and justice policies based on perceived attitudes (Adriaenssen & Aertsen, 2015; Roberts et al., 2002).

Nevertheless, there is a scarcity of studies exploring the public's views on RJ in cases of sexual abuse. An Irish study (Keenan & Zinsstag, 2014) revealed public willingness to engage in RJ in sexual abuse cases. Similarly, research in England (Marsh & Wager, 2015) showed that 81% of participants and 70% of self-identified victims supported victim-offender encounters in sexual offense cases. More recently, a study conducted in Israel found that the majority of the public supports the idea of RJ in sexual abuse cases and understands the advantages inherent in the process (Peleg-Koriat & Klar-Chalamish, 2020).

These groundbreaking studies highlight the public's favorable attitude toward RJ in such contexts. However, these studies have not examined the differences in public attitudes toward RJ processes in cases of intrafamilial versus extrafamilial sexual abuse of minors, nor have they considered ethnic differences—the twin focuses of the current research.

Attitudes Toward Restorative Justice in Intrafamilial and Extrafamilial Sexual Abuse

Sexual abuse of children and adolescents is a uniquely traumatic experience that continues to resonate throughout their lives (Klar-Chalamish & Peleg-Koriat, 2021; Sánchez-Meca et al., 2011). Preventing and addressing these abuses requires a multidisciplinary approach, with substantial efforts and coordination among public administrators, numerous professionals, families, victims, and society at large. Numerous studies indicate that abuse can manifest in various forms and have different consequences, depending on whether the incidents occur within or outside the family (Hartley & Bartels, 2022; Magalhães et al., 2009).

The literature indicates that perpetrators of extrafamilial sexual abuse are typically more dangerous and pedophilic than perpetrators of intrafamilial abuse (Seto et al., 2015). However, according to research on incest taboo (Antfolk et al., 2012), the public considers the acts of intrafamilial perpetrators as more severe, a perception likely due to their incestuous nature, as suggested by the Westermarck effect (Westermarck, 1891). Hence, the current assumption is that in general, people would be less supportive of RJ processes in cases of sexual offenses within the family.

Reynolds and Birkimer (2002) demonstrated that the disparities between an external and an internal family perpetrator increased in the case of adolescent girls. They found that an 8-year-old girl was perceived as highly abused regardless of whether the abuser was a stepfather or a neighbor. In contrast, when the victim was 15 years old, she was considered less abused if the perpetrator was a neighbor rather than a stepfather. Similarly, Davies and Rogers (2009) found that child sexual abuse was regarded as more severe when committed by the victim's father compared to a stepfather. Conversely, Hartley and Bartels (2022) found that in extrafamilial cases (stranger and babysitter), the perceived danger and pedophilic tendencies were higher compared to a stepfather preparator.

The identities of the participants, their basic beliefs, and the offender–victim relationship may explain these unique and contradictory findings. Further research is required to thoroughly investigate the psychological mechanisms influencing the perceived severity of sexual offenses within and outside the family, as well as public willingness to support RJ processes in these cases, a topic that yet to be empirically studied. This study contributes to that end by examining, for the first time, attitudes toward RJ in cases of both intrafamilial and extrafamilial sexual abuse within a multiethnic society. Based on the literature reviewed above, indicating that differences in public perceptions emerge mainly in adolescence, it focuses on adolescent victims. Additionally, it investigates the mediating role of perceived offense severity and malleability beliefs in the relationship between the offense context (intra- or extrafamilial) and support for RJ.

Perceived Severity of Sexual Offenses, Malleability Beliefs, and Support for Restorative Justice

As explained above, the current study posits that although there is no definitive answer in the literature, the public is likely to perceive sexual offenses within the family as more severe. It is further hypothesized that this perception would negatively influence the public's belief in the offender's ability to change (malleability beliefs), which in turn will adversely affect support for RJ in cases of intrafamilial sexual offenses.

Dweck et al. (1995) propose that individuals hold one of two beliefs regarding the malleability of personal character traits: they either view these traits as unchangeable (entity theory) or as changeable (incremental theory). Research across various fields has demonstrated that holding incremental beliefs is linked to a reduced likelihood of making stereotypical judgments (Rydell et al., 2007), a decrease in negative reactions to social adversity (Yeager et al., 2014), and lower tendency to endorse punishment and retaliation in response to others’ misconduct (Chiu et al., 1997). Individuals who subscribe to incremental beliefs are generally inclined to perceive negative behaviors as challenges, and they are more likely to propose constructive solutions aimed at addressing the underlying motivations or circumstances. In contrast, those who believe that wrongdoing stems from unchangeable traits tend to favor punitive measures and retaliation (Dweck et al., 1995).

Recently, the exploration of incremental beliefs within the context of criminology has gained attention. For instance, Tam et al. (2013) discovered that individuals with entity beliefs were more prone to make internal attributions for criminal behavior, leading to stronger punitive inclinations. Rade et al. (2018) found that incremental beliefs predicted positive attitudes toward ex-offenders, fostering greater public support for their reintegration. Finally, Harper and Bartels (2017) found that people with entity-based beliefs made more punitive judgments about sexual offenders compared to those with incremental beliefs.

Specifically in the context of RJ, Moss et al. (2019) and Peleg-Koriat et al. (2020) found that participants who believed offenders could change were more likely to support RJ and rehabilitation over punishment. Moreover, Weimann-Saks et al. (2022) suggested that the greatest challenge in altering public attitudes toward RJ would be in the case of individuals who have committed sexual offenses given their perceived low malleability.

Accordingly, the current study hypothesizes that sexual offenses within the family would be perceived as more severe, leading to a stronger belief that the offender cannot change, and resulting in weaker support for RJ processes. Moreover, the current study seeks to present a more complex picture and examine whether this mediation model is applicable to both the Jewish and Arab communities in Israel.

Ethnicity and Attitudes Toward Restorative Justice in Sexual Abuse

The present study assumes that attitudes toward RJ would also be affected by the participant's ethnicity. Specifically, differences are expected between Arabs and Jews in Israel in their attitudes toward RJ in cases of both intrafamilial and extrafamilial sexual abuse, which stem from the participants’ culture.

Israeli society is made up of multiple sociocultural groups, the largest being the Jewish majority and Arab minority (21.1% of the total population; Central Bureau of Statistics, 2022). The Jewish majority is characterized by strong pronatalist and familial attitudes. Nevertheless, Jewish society is largely secular, individualistic, and Western (Sharabi et al., 2020), with gender patterns and attitudes resembling those in other industrialized societies (Kulik et al., 2016; Lavee & Katz, 2003). Therefore, we will not elaborate on its characteristics here.

Arab culture, on the other hand, is more religious and marked by collectivism. In Arab families, relationships are typically hierarchical, age-based, and patriarchal, with children expected to be submissive to their parents (Abu-Baker, 2013; Attrash-Najjar & Katz, 2023; Haboush & Alyan, 2013). The traditionalism of Arab society also means an emphasis on modesty and restrictions on sex education, inhibiting open conversations about sexual abuse. The situation is further complicated when the perpetrator is a family member, as Arab families struggle between loyalty to the offending relative and the need to protect the victim, leading often to the decision to keep such incidents hidden (Abu-Baker, 2013).

Responses to sexual abuse in Arab societies, worldwide as well as in Israel, face several challenges, including the taboo nature of the topic and the difficulty in accessing child victims (Attrash-Najjar & Katz, 2023). In some instances, families view the victim as a threat to the family's honor and reputation (Abu-Baker, 2013; Hasan, 2003). In particular, loss of virginity is considered a source of shame for the entire family, leading to a cultural norm where children are taught to avoid discussing sexuality (Attrash-Najjar & Katz, 2023; Shalhoub-Kevorkian, 2005).

Arab families often prefer to handle criminal matters within the community or family, such as through the traditional sulha process. Sulha (“reconciliation”) is a common practice with religious roots, similar to RJ. It aims to transform the parties’ attitudes from the desire for retribution to forgiveness (Pely & Luzon, 2018). This preference stems, among other things, from concerns that involving authorities might damage the family's reputation. In the case of Arab society in Israel, this apprehension is rooted in a widespread sense of distrust and alienation toward state-provided mental health services and the discriminatory criminal justice system (Al-Krenawi, 2009; Diab & Sandler-Loeff, 2011). Studies have shown that Arabs in Israel report lower levels of trust in the police compared to Jews, which affects their willingness to report crimes (Ben-Porat & Yuval, 2012; Hasisi & Weisburd, 2014; Nanes, 2020).

Informed by this literature, this study hypothesizes that RJ, which is characterized by privacy and integration, would be well-received in Arab society, particularly in cases of offenses considered “shameful.” Furthermore, in contrast to the individualistic Jewish society, in Arab society where sexual offenses within the family are seen as tarnishing both the victim and family and discretion is more valued, there would be support for RJ even in these cases.

The Present Study

The present study examined differential public attitudes toward RJ processes in cases of intrafamilial versus extrafamilial sexual assault against adolescents. The purpose was to investigate attitudes toward RJ processes between an offender and an adult victim, where the offense occurred when the victim was a minor. It tests a comprehensive model that integrates a two-stage mediation of offense severity and malleability beliefs, moderated by ethnic identity. The model is intended to elucidate the underlying psychological and cultural mechanisms that shape these attitudes. By employing a rigorous quantitative approach, this study addresses a significant gap in the literature and provides a nuanced perspective on the complexities of public perceptions regarding RJ in the context of sexual violence. Specifically, this research examines the following hypotheses:

H1: Arab participants would support RJ more than Jewish participants would in cases of sexual abuse, whether intra- or extrafamilial.

H2: Participants would judge intrafamilial sexual abuse as more severe than extrafamilial sexual abuse.

H3: Based on a serial mediation model wherein intrafamilial sexual abuse is associated with higher offense severity compared to extrafamilial sexual abuse, higher offense severity would be linked to lower perceived malleability, which in turn, correlates with reduced support for RJ.

H4: The mediation effect would be moderated by ethnicity. Specifically, while significant mediation is expected in both ethnic groups, the effect would be stronger among Jews.

In addition to the specific research hypotheses, participants were asked about their support for retributive punishment, as well as whether they believed punishment should consider the victim's needs. These questions were examined separately from the primary hypotheses to preliminarily assess whether participants support RJ as opposed to punitive retributive processes, and to explore the potential relationship between support for RJ and the desire to consider the victim's needs, or vice versa.

Method

Participants

The sample size in Study 1 was determined using GPower (Faul et al., 2009), based on small- to medium-sized effects (Cohen's f = .18), to obtain a 95% power to detect significant differences. The sample was obtained from iPanel, a major and reliable Israeli online panel. Since 2006, iPanel has provided an online platform for various information collection services, including polls and public opinion surveys. Participants sign up to the panel and get paid for their participation (the equivalent of two dollars for this study). We used quota sampling to obtain a roughly even number of Jews and Arabs. The study's 517 participants included 257 Jews and 260 Arabs.

Demographic characteristics and group comparisons are presented in Table 1. Jewish participants were significantly older. There was also a significant difference in marital status, and post hoc analysis (with Bonferroni corrected p values) showed Jews had a lower proportion of single participants, χ²(1) = 8.28, p = .012. No difference was observed in the “married,” χ²(1) = 4.31, p = .111, or “other” category, χ²(1) = 1.51, p = .654. Jews also had a higher income status, but both groups had similar education levels and gender distributions.

Table 1.

Demographic Characteristics.

	Jews (n = 257)	Arabs (n = 260)	Statistics	p
Age	46.15 ± 13.99	37.66 ± 12.32	t(505.39) = 7.30	<.001
Education	4.42 ± 1.30	4.26 ± 1.30	t(514.92) = 1.31	.189
Income status	3.09 ± 1.20	1.99 ± 1.13	t(468) = 10.16	<.001
Gender (% female)	49.41 (n = 127)	50.00 (n = 130)	χ2(1) = .01	.894
Marital status (%)	Single—24.51 (n = 63)	Single—36.15 (n = 94)	χ2(2) = 8.79	.012
	Married—66.92 (n = 172)	Married—58.07 (n = 151)
	Other—8.56 (n = 22)	Other—5.76 (n = 15)

Measures

All measures were assessed on a six-item scale, from 1 (strongly disagree) to 6 (strongly agree).

Offense severity (α = .736) was assessed based on Warr's (1989) and Alter et al.'s (2007) questionnaires, with minor adjustments. The questions identified the wrongfulness of criminal acts in terms of perceived seriousness (e.g., “The behavior described deviates from accepted norms”) and harm (e.g., “The behavior described may cause serious injury”).

Malleability beliefs (α = .916) were assessed based on the original scale by Chiu et al. (1997), with minor modifications by Weimann-Saks et al. (2019). These adjustments were tailored specifically to the domain of the current research (e.g., “I believe that a person who has committed such an offense has a chance to return to a normative life”; “I think a person who commits such an offense can change”).

RJ support (α = .928) was assessed by a scale based on Peleg-Koriat and Klar-Chalamish's (2020) questionnaire, with slight alterations (e.g., “I think it is right that the victim will meet the person who harmed him as part of an RJ process”; “I welcome the opportunity given to the victim and offender to choose whether they are interested in an RJ process”).

Retributive punishment support (α = .814) was assessed by a scale based on Einat and Lallo's (2011) questionnaire, with slight alterations (e.g., “George needs to be sent to prison”; “It is fair for society to punish people like George who commit crimes against it”).

Victim-oriented punishment (α = .893) was assessed by a scale developed for the current study (e.g., “Anna's well-being should be considered when considering George's punishment”; “Anna's wishes should be paramount in George's punishment considerations”).

Procedure

Participants were assigned to one of two conditions using the online panel's built-in randomization mechanism, one detailing an intrafamilial and the other an extrafamilial sexual abuse case. The victim and offender were named Anna and George, respectively, rather than using popular local names, to prevent participants from assuming the characters’ ethnicity, thereby minimizing the potential of racial bias or prejudice. In the two conditions, the victim was an adult woman who pressed charges against a man who had abused her when she was 16. In the intrafamilial condition, the man was her uncle, and in the extrafamilial condition, the man was an employee at the community center (see Appendix A for the full scenarios).

After reading the scenarios, the participants were then asked to complete a questionnaire regarding their stance on the severity of the offense and their malleability beliefs. After completing these two scales, participants read a short description of the RJ procedure (see Appendix B, adapted from Peleg-Koriat and Klar-Chalamish, 2020), and were asked to complete a questionnaire that examined their support for RJ, retributive punishment, and victim-oriented punishment. Approval for the present study was granted by the institutional ethics committee.

Data Analysis

Before factor analysis was conducted, sampling adequacy for the newly developed victim-oriented punishment questionnaire was examined using the Kaiser–Meyer–Olkin (KMO) test, and sphericity was examined using Bartlett's test. The KMO index was .862, suggesting the sampling was adequate for factor analysis. Bartlett's test was significant, χ²(15) = 1878.97, p < .001, suggesting the correlation matrix was not an identity matrix, and was therefore suitable for factor analysis. Next, factor analysis was conducted and yielded a single factor (eigenvalue = 3.94), which explained 65.82% of the variance. Factor loadings were: Q1- .737, Q2- .816, Q3- .837, Q4- .873, Q5- .834, Q6- .759.

After calculating Pearson correlations, the first and second hypotheses were examined via independent samples t-test and ANCOVA, respectively. The third and fourth hypotheses were examined via structural equations analysis (AMOS v27). Bootstrapping (5,000 samples) was conducted to assess the significance of the model coefficients.

Results

Preliminary Analyses

The preliminary analyses examined whether people viewed RJ as contradictory to retributive justice and whether support for RJ was associated with attitudes prioritizing victim-centered considerations in the context of punishment. To do so, Pearson correlations between support for RJ, retributive punishment, and victim-oriented punishment were calculated. Negative and significant correlations were found between RJ support and retributive punishment support (r = −.493, p < .001; 95% bootstrapped CI [−0.558, −.0426]); and between RJ support and victim-oriented punishment (r = −.207, p < .001; 95% bootstrapped CI [−299, −.113]). These correlations suggested that higher support for retribution, and higher consideration of the victim regarding the offender's punishment, were correlated with lower RJ support. Note that the correlation of RJ support with retributive punishment support was significantly higher than the one with victim-oriented punishment (Z = −5.29, p < .001; Fisher z-transformation).

Since group differences were found in age, income, and marital status, it was important to examine whether they correlate with any of the variables. As age was correlated with RJ support for Jews (but not Arabs), it was used as a covariate in all analyses examining ethnicity differences. Income and marital status were not correlated with any of the study's variables, so they were excluded from the analyses.

Hypothesis Testing

The first hypothesis suggested Arab participants would support RJ more than Jewish participants in cases of sexual abuse, whether intra- or extrafamilial. A significant ANCOVA test, F(1, 514) = 9.76, p = .001, adjusted for age, revealed that as hypothesized, Arab participants (3.23 ± 1.24) supported RJ more than did Jewish participants (2.96 ± 1.20).

According to the second hypothesis, participants would judge intrafamilial sexual abuse as more severe. A significant independent samples t-test, t(514.18) = 2.11, p = .035, confirmed the hypothesis: participants found intrafamilial sexual abuse (6.18 ± 0.98) to be more severe than extrafamilial sexual abuse (5.99 ± 1.03).

The third hypothesis suggested a serial mediation model, wherein intrafamilial sexual abuse was associated with higher offense severity compared to extrafamilial sexual abuse. As higher offense severity is linked to lower perceived malleability, this would correlate with reduced support for RJ. Additionally, the mediation effect was hypothesized to be moderated by ethnicity (H4), so that the effect would be stronger among Jews (Table 2).

Table 2.

Moderated Mediation Model Analysis of RJ Support Through Intra-/Extrafamilial Sexual Abuse, Offense Severity, and Malleability.

	Path	B	95% CI (Bootstrapped)	p
Jews	Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Offense severity	.268	.090, .446	.002
	Offense severity → Malleability	−.328	−.541, −.127	.001
	Malleability → RJ support	.226	.124, .331	<.001
	Direct effect: Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ RJ support	.020	−.253, .294	.870
	Indirect effect: Intra/Extrafamilial Sexual abuse (1 = Intrafamilial)→ Offense severity → Malleability → RJ support	−.020	−.052, −.006	.001
	Additional paths
	Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Malleability	.122	−.206, .426	.470
	Offense severity → RJ support	−.356	−.526, −.173	<.001
Arabs	Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Offense severity	.054	−.222, .328	.702
	Offense severity → Malleability	−.219	−.334, −.104	.001
	Malleability → RJ support	.124	−.041, .289	.138
	Direct effect: Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ RJ support	−.022	−.324, .258	.850
	Indirect effect: Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Offense severity → Malleability → RJ support	−.001	−.018, .005	.464
	Additional paths
	Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Malleability	−.209	−.507, .059	.128
	Offense severity → RJ support	−.259	−.389, −.129	.001
Moderated mediation		−.018	−.050, −.001	.034

Note. Age was used as a covariate. RJ = restorative justice.

The hypothesis was partially confirmed (see Figure 1 and Table 2). For Jews, but not Arabs, intrafamilial sexual abuse was associated with higher offense severity. In both groups, higher offense severity was associated with lower malleability, but only for Jews were higher malleability beliefs associated with higher RJ support. The direct effect in both groups was nonsignificant. A negative and significant serial mediation was found for Jews, but not for Arabs, so that for Jews, intrafamilial sexual abuse was associated with lower RJ support, compared to extrafamilial sexual abuse. This effect was fully mediated by offense severity and malleability beliefs. The strength of the mediation was significantly moderated by ethnicity, so that the mediation effect was significantly higher for Jews compared to Arabs (whose mediation effect was nonsignificant).

Figure 1.

Serial Moderated Mediation Model of RJ Support Through Intra- and Extrafamilial Sexual Abuse, Offense Severity and Malleability. Note. *p < .05, **p < .01, ***p < .001.

The full model was saturated, so that model fit indices could not be computed. To estimate the model's goodness of fit, the weakest path of the model that did not involve a covariate [Intra-/Extrafamilial sexual abuse (1 = Intrafamilial)→ Malleability] was removed. The path was removed for this analysis only and is included in the full model. All path coefficients remained virtually the same, and high goodness of fit was found: χ²(2) = 2.72, p = .256; χ2 relative = 1.36; CFI = .992; TLI = .924; RMSEA = .027, p = .606; SRMR = .012; Bollen-Stine bootstrap p = .263. These goodness-of-fit values suggest a well-specified model and a high degree of compatibility between the theoretical model and data.

Discussion

As public attitudes play a crucial role in the implementation of RJ practices (Adriaenssen & Aertsen, 2015; Indermaur et al., 2012), their comprehensive understanding is important. Thus, the primary aim of the present study was to explore the public attitudes toward RJ in cases of sexual abuse, with a specific focus on the distinction between intra- and extrafamilial sexual abuse. Additionally, this research sought to investigate, for the first time, whether ethnicity—specifically, Arabs and Jews in Israel—influenced these attitudes.

A key preliminary finding is that ethnic identity significantly influences support for RJ processes in cases of sexual offenses. Specifically, Arabs exhibited greater support for RJ than did Jews. This finding can be understood from a cultural perspective and in light of the sociopolitical status of Arabs in Israel. In Arab culture, family honor and communal cohesion are paramount, often leading to a preference for maintaining the family's reputation over seeking justice through a formal legal system (Haj-Yahia & Tamish, 2001; Shalhoub-Kevorkian, 2005). The intense concern for family honor and fear of stigmatization can result in victims avoiding disclosure of abuse (Shalhoub-Kevorkian, 2005), and sometimes have extreme consequences such as “honor killings.” A Lebanese study revealed that among the 54% of children who disclosed sexual abuse to their parents, responses varied widely, from carefully listening to dismissing the disclosure or instructing the child not to discuss it (Usta & Farver, 2010).

This cultural context suggests that RJ may be particularly effective, as it offers a conflict resolution approach that aligns with the cultural emphasis on safeguarding the family and victim's honor. The current study proposes that RJ can align with these cultural emphases, providing a framework that may be more acceptable within the societal norms of the Arab community. This preference is consistent with Braithwaite's (1989) reintegrative shaming theory. In cultures where avoiding public disgrace is critical, as in Arab society, reintegrative shaming offers a culturally sensitive framework that supports RJ, avoiding the stigmatization associated with formal legal procedures.

Another explanation for the greater support of Arabs in Israel for RJ is related to their underrepresentation in and discrimination by the police force and judicial system. These factors likely drive the inclination toward RJ, which can operate independently of the state. Comparative studies reveal that Arabs exhibit significantly lower trust in the police than do Jews, which adversely impacts crime reporting (Ben-Porat & Yuval, 2012). For instance, Hasisi and Weisburd (2014) found that while 70.1% of Jews would contact the police if they witnessed a crime, only 50.6% of non-Jews would do so (see also Nanes, 2020). This highlights the need for additional conflict resolution procedures in Israel as a whole and among Arabs in particular. RJ, in these cases, should not be presented to the public as an alternative to traditional law enforcement, but rather as an alternative to something more common—disregarding the harm and offering no intervention at all (Peleg Koriat & Klar Chalamish, 2022).

In a more global context, the present findings underscore the significance of cultural factors in research on attitudes toward RJ. In general, it was found that in cases of intrafamilial sexual abuse, there was lower support for RJ compared to extrafamilial sexual abuse. This is consistent with the current hypothesis that the public perceives intrafamilial sexual abuse as more severe and views the offender in these cases as undeserving of “leniency,” in the form of participation in RJ. The preliminary findings of this study—a negative correlation between prioritizing victim-centered considerations in sentencing and support for RJ, and the negative correlation between support for retributive punishment and support for RJ—suggest that the public perceives RJ as an offender-focused process that offers a degree of leniency to the perpetrator. The study's findings show that this perceived leniency is a “privilege” the public is unwilling to extend to those who commit sexual offenses within the family. However, a deeper analysis of the findings and the research model reveals a more nuanced picture that incorporates cultural factors.

Among Jews, intrafamilial sexual abuse was perceived as more severe than, leading to the belief that the offender was incapable of change and thereby unsuitable for participation in RJ processes. In contrast, among Arabs, no significant difference was identified in the perceived severity of intra- and extrafamilial sexual abuse. Additionally, among Arabs, belief in the offender's malleability was not associated with support for RJ. This finding is particularly significant as it contradicts a substantial body of recent research, which has consistently highlighted malleability beliefs as a crucial variable in fostering support for RJ (e.g., Moss et al., 2019; Peleg-Koriat et al., 2020). According to Weimann-Saks et al. (2022), belief in an offender's malleability is often more influential in generating support for RJ than is the perceived severity of the offense or fear of the offender, even in the context of sexual offenses. The current study introduces a new perspective by showing that the impact of belief in malleability on support for RJ is culturally dependent, varying between Jews and Arabs in the context of sexual offenses. The collectivistic nature of Arab culture and the significance attributed to the honor of the family and close community (Abu Baker, 2013; Attrash-Najjar & Katz, 2023; Haboush & Alyan, 2013) may be more influential than fundamental psychological beliefs regarding the offender's nature in shaping attitudes toward RJ.

Another significant finding is that for the Jewish participants, the primary challenge is supporting RJ in intrafamilial sexual abuse cases, even though previous studies have demonstrated the importance of such processes precisely in those cases. A recent study by Klar-Chalamish and Peleg-Koriat (2021), relying on interviews with incest survivors, nonoffending family members, friends, and RJ facilitators, highlighted the unique contribution of RJ to families affected by sexual abuse. Participants described the RJ process as pivotal in their recovery by establishing agreements on future conduct within the family. They viewed RJ as addressing unique needs in incest cases within a supportive environment. These findings align with the literature highlighting RJ's benefits in facilitating recovery and growth in both intra- and extrafamilial sexual offenses (Jülich, 2006, 2010; McGlynn & Westmarland, 2019).

In sum, understanding the cultural and psychological mechanisms that shape public attitudes is essential for advancing and regulating RJ processes tailored to diverse populations in sexual abuse cases. The management of these processes is crucial: they must be conducted with the utmost professionalism and sensitivity, ensuring that victims feel a sense of control and are fully and respectfully heard. Prioritizing the victim's protection and well-being, while also addressing the needs of all participants, are paramount.

Limitations and Future Directions

Despite its contributions, the present study has several limitations. First, it differentiates between intrafamilial and extrafamilial sexual abuse using two conditions. In one, the perpetrator is the victim's uncle, while in the other, the perpetrator is employed at a community center. It is reasonable to assume that varying types of relationships—whether more distant (e.g., assault by a stranger) or closer (e.g., assault by a brother or father)—could affect the findings, warranting examination in future research.

Second, the respondents were not informed or given any indication of the ethnic identities of the perpetrator and victim in order to avoid ingroup bias (Tajfel, 1982). This bias results in more lenient judgments when the offender is perceived by the participant as part of the same group (Schiller et al., 2014; Wenzel & Thielmann, 2006). Future research should explore the interaction between participants’ ethnicity and that of the parties involved in the crime.

Third, the current study focused on an adolescent victim. Prior research suggests that while intrafamilial abuse is perceived as especially severe regardless of age, due to the incest taboo, and while abuse of young children is viewed as unequivocally serious, cases involving adolescents introduce additional considerations, some of which pertain to the issue of consent (Reynolds & Birkimer, 2002). However, the findings indicate that this pattern is not uniform across all groups, as Arab respondents did not perceive intrafamilial abuse as more severe than extrafamilial abuse. To achieve a deeper understanding of this finding, further research should examine cases involving victims and perpetrators of different ages, from different cultural backgrounds, and in instances where the victim and perpetrator come from different ethnic groups.

Last but not least, the Arab participants in this study included both Muslims and Christians. Christians constitute 7.1% of the Arab population in Israel (Israel Central Bureau of Statistics, 2020) and thus represent a minority religion within a minority ethnicity (Raheb, 2002). In the current study, given that the Christians represented a minority similar to their proportion in the general population of Israel, it was not feasible to examine them as a separate group. Similarly, the Jewish population is not monolithic. The Haredi (ultra-Orthodox) community, a minority within the Jewish population, shares certain values with Arab society, such as an emphasis on family honor and lack of cooperation with state authorities, which affect disclosure in the case of sexual abuse (Tener & Katz, 2021). The research literature will also benefit from further studies exploring intra-group differences in attitudes toward RJ among various religions within Arab society.

Hopefully, the present study will inspire further research in this area, as well as the continued implementation of RJ processes in cases of intrafamilial and extrafamilial sexual abuse, as well as in others.

Author Biography

Inbal Peleg-Koriat, is a lawyer, senior lecturer, and faculty member in the Department of Criminology at the Yezreel Valley Academic College. She earned her PhD from the Conflict Resolution, Management and Negotiation Graduate Program at Bar-Ilan University, Israel. Her research focuses on therapeutic jurisprudence and restorative justice, particularly in prison settings and in cases involving sexual victimization.

Appendix A: Questionnaire Scenarios

Intrafamilial Condition

On June 24, 2023, Anna filed a complaint with the police against George for indecent acts committed against her when she was 16 years old. According to Anna's testimony, her uncle George, her father's brother, performed indecent acts on her on several different occasions. The abuse Anna described included touching her intimate parts when she was alone with George in the family home. Anna said that during the moments of abuse, she felt she was “freezing in place,” unable to react or resist. Anna was only able to talk about the incident after two years.

Extrafamilial Condition

On June 24, 2023, Anna filed a complaint with the police against George for indecent acts committed against her when she was 16 years old. According to Anna's testimony, George, who worked at the community center in her town, performed indecent acts on her on several occasions. The abuses Anna described included touching her intimate parts while she was with George at the community center. Anna said that during the moments of abuse, she felt she was “freezing in place,” unable to react or resist. Anna was only able to talk about the incident after two years.

Appendix B: Description of Restorative Justice

Restorative justice is an approach from the field of criminal law that focuses on the needs of perpetrators and victims, rather than the need to punish the offender to the letter of the law. The main goal of the restorative justice procedure is to redress the harm caused to individuals as a result of a criminal offense. This goal is achieved through a guided, fair dialogue between the parties affected, which is similar to the mediation process, with one crucial difference: the parties are not equal as their identities as perpetrator and victim are clear. Healing the vulnerability is enabled when the offender recognizes the act and the harm as well as the victim's consequent needs, accepts responsibility, and acts to meet them (for example, by providing financial compensation, apology, and community service). The process takes place only with the consent of the parties involved and can either complement or replace the legal process.