The Richness of Reports Provided by Children in Israel: Examination of the Impact of Social Affiliations

Abstract

Forensic interviews following child sexual abuse (CSA) are of central importance to the children, their families and all those involved. Moreover, the legal system expects rich, forensically relevant reports from the children. The current study focuses on the impact of children’s social affiliation on the richness of their reports, and how question types contribute to rich reports. The sample included 314 forensic interviews conducted in Israel between 2015 and 2018. The findings revealed a relationship between child characteristics (gender), abuse characteristics (perpetrator identity, abuse type, abuse frequency) and social affiliation with report richness. Furthermore, question types (free recall prompts, summaries, directive, option-posing, suggestive) moderated the relationship between the child’s characteristics, abuse characteristics, and social affiliation with report richness, when these effects were not equal. The findings emphasized that contextual observation of sexually abused children may promote better services for them and, in addition, stressed the importance of advancing future training and practical guidelines for practitioners.

Keywords

forensic interviews Arab Muslim children Jewish children child sexual abuse richness of report question types

Introduction

Child sexual abuse (CSA) is a widespread social issue causing significant harm to survivors (Malloy et al., 2011). Despite its prevalence, understanding CSA is limited due to underreporting, which obscures its full scope across different social and cultural contexts (Nadan et al., 2015). In Israel, forensic interviews are crucial for obtaining detailed information following CSA disclosure, utilizing various question types (Lamb et al., 2011), following the NICHD protocol. This study addresses the relationship between social affiliation and the richness of reports (RR) in forensic interviews of sexually abused children, focusing on Arab Muslim and secular, religious, and ultra-Orthodox Jewish children.

Social Affiliation

The current study delves into the multifaceted nature of social affiliations, encompassing cultural, social, and religious dimensions that play a pivotal role in children's lives within diverse societies (Tishelman & Fontes, 2017; Nadan & Ganz, 2018). Understanding how these social affiliations influence forensic interviews is essential for fostering comprehensive and culturally sensitive investigative practices. By exploring the impact of social affiliations on children's RR in forensic interviews within Israel's multicultural society, including Jewish and Arab populations with varying religiosity levels, this study sheds light on the intricate interplay between social affiliation and RR. While the research focuses on specific social groups in Israel, the insights gained are intended to offer valuable knowledge that can be extrapolated to diverse cultural and social contexts worldwide, underscoring the significance of adopting culturally informed approaches in investigative procedures.

Within Israel’s diverse social landscape, secular Jewish society constitutes 45.3% of the Jewish population (Central Bureau of Statistics [CBS], 2021). Most of this society is characterized as individualistic, often placing the child at the center of the family and affording the child liberal values of separation and independence (Ben-Porat et al., 2022). Accordingly, parents often give their children relative freedom to make and take responsibility for their decisions. Thus, there is a large measure of liberalism, with the preservation of the individuality and uniqueness of the child within the family context (Ben-Porat et al., 2022; Shoham, 2014).

Religious Jewish society comprises 43.8% of Israel’s Jewish citizens (CBS, 2021). It includes many streams that are not uniform in patterns of religious behavior (Shoham, 2014). Nevertheless, this society is considered, both in its own eyes and in those of Israeli society in general, to be a distinct group (Shoham, 2014). The modern reality is especially complex due to this society’s double ideological commitment to the system of Jewish law (halacha) and the State and its liberal ways (Khoury, 2013). This often creates ideological and pedagogical difficulties, particularly when a conflict exists between the two systems (Sharabi & Kay, 2021). In recent years, there has been growth in the “area of friction” between religious Jewish society and the broader society (Khoury, 2013; Sharabi & Kay, 2021). This friction has generated a sense of entitlement to influence the lifestyle of Israeli society while also adopting various influences of the modern Israeli world (Sharabi & Kay, 2021). Accordingly, religious Jewish society generally cultivates values of self-actualization and achievement-oriented thinking among their children. These values also include a commitment to family and social norms, alongside behaviors and actions that create and shape identity and are likely to create differences between them and other children (Sharabi & Kay, 2021).

In contrast, ultra-Orthodox Jewish society, which is 10.5% of Israel’s Jewish citizens (CBS, 2021), tends to sanctify the values of family and collectivism alongside clear boundaries and isolation from the overall society (Nadan et al., 2019; Poniscjakova, 2021). The ideology and lifestyle of this society position it in opposition to the greater surrounding society (Cahaner, 2023; Keidar et al., 2022). Some of the unique expressions include its perception of Jewish life, the way society and its institutions are conducted, the school system and acceptable dress (Cahaner, 2023). The religious ethos in this society preaches solidarity and collectivism. In this context, the Jewish worldview permits and even obliges intervention in the lives of others, including children (Gemara & Nadan, 2022). The parent-child relationship in this society is primarily characterized by the requirement of children to conform and obey social norms, and a severe attitude towards those who deviate from the social order and are sexually permissive and hedonistic (Keidar et al., 2022).

Arab Muslim society is an ethnic minority in Israel, constituting 15% of citizens (CBS, 2021). This society is characterized mainly by a traditional collective orientation, mutual dependence, and family and social cohesiveness (Fahoum & Al-Krenawi, 2023). Additionally, it is undergoing a process of “selective modernization” due to everyday contact with secular Jewish Israeli society, through which the individualistic characteristics of the majority are imprinted (Jaraisi, 2012; Nijim-Ektelat et al., 2018). The social and cultural changes in Arab Muslim society include the reduced influence of the clan and extended family as well as changes in the structure and role assignments within the family, including the status of children, which greatly contributes to conflicts related to raising and educating children (Al-Amer et al., 2024). At the same time, the extended family continues to play a central role in the lives of Arab Muslim children and youth (Al-Amer et al., 2024). There is still a tendency to expect children to turn to their parents or other family members for help during difficulties. Likewise, the social, cultural, and socio-political relationships make it difficult for children to use formal services for fear of stigma and sharing their problems with strangers (Al-Amer, 2024; Jaraisi, 2012; Nijim-Ektelat et al., 2018). When flouting traditional norms and non-acceptance of the cultural and social values of the family, it is likely that a child will not receive support from their family, but rather social rejection and even punishment (David & Khatib, 2021).

The existing literature underscores the importance of considering the social affiliations of abused children to gain a comprehensive understanding of their circumstances. These affiliations encompass religious and socio-political contexts that are intricately intertwined and challenging to disentangle (Mehrotra, 2010). While studies have explored the impact of social affiliation on different religious communities, such as Catholics (Parkinson, 2014), Amish (McGuigan & Stephenson, 2015), and Mormons (Gerdes et al., 1996), there remains a scarcity of empirical research on CSA reports within multicultural settings. Therefore, it is imperative to delve into how a child’s social affiliation contributes to RR in cases of abuse, emphasizing the need for further investigation and understanding in this critical area.

Richness of Reports

The concept of RR concerns the number of details provided by abused children during forensic interviews (Lamb et al., 2011; Malloy et al., 2011). The legal system anticipates rich, forensically relevant reports from children, which are crucial for evidence (Lamb et al., 2011). However, this richness is influenced by many variables, including the child’s characteristics and the abuse characteristics.

Regarding the child’s characteristics, studies have indicated that older children tend to provide richer reports due to enhanced cognitive and linguistic abilities, which affect memory retrieval, vocabulary, and the capability to express themselves (Hershkowitz et al., 2014; Piaget, 1936/1952). Some studies have indicated that following the NICHD protocol can minimize gender effects on RR (Brown et al., 2019; Leach et al., 2017). However, societal pressures may lead boys to provide less detailed reports, while girls may be more open about their emotions, potentially influencing RR (Alaggia et al., 2019; Easton et al., 2014). Concerning the abuse characteristics, studies have attested to a relation between RR and perpetrator identity. Providing reports implicating familial perpetrators often involves emotional difficulties and conflicted loyalties (Leach et al., 2017; Malloy et al., 2011), leading to fewer details compared to reports of CSA by non-family members (Leach et al., 2017).

The association between severe and frequent abuse with less detailed reports may be attributed to how these traumatic events are encoded in memory. Studies have suggested that repeated trauma can lead to suppression and storage outside the child’s consciousness, potentially explaining the association with less detailed reports (Freyd, 1996; Leach et al., 2017). In contrast, a one-time trauma is typically more readily accessible for memory retrieval. This difference in memory processing, coupled with the challenges of recalling detailed information in cases of recurring trauma, could contribute to the scarcity of evidence in such instances (Korkman et al., 2014; Lamb et al., 2011).

Only a few studies have discussed the relation between RR and the social affiliation of abused children (Hindi et al., 2022; Mordi et al., 2022). According to Hindi et al. (2022), children from more individualistic societies provide richer reports than those from more collectivistic societies. This difference is due to divergent social and family norms influencing their perceptions, understanding and attitudes toward CSA (Nadan et al., 2015).

Types of Questions Used in Forensic Interviews

The forensic interview’s role is to encourage children to provide narratives that advance the legal decision-making process (Katz & Barnetz, 2014). Many studies have addressed effective interviewing strategies to obtain a report from a child that is of high quality, clear and credible (e.g., Malloy et al., 2011), delineating a typology of questioning styles. This classification includes a distinction between free recall prompts (“Tell me everything that happened”) and cued prompts (“Earlier you mentioned a man. Tell me everything about him”). There are also direct questions (“Who touched you? How did they touch you?”), option-posing questions (“Did he touch you on top of your clothing?”, relevant when the child has previously mentioned it), and suggestive questions (“Tell me how he touched you on top of your clothing”, asked when the child has not mentioned being touched on top of their clothing).

Findings have indicated that free recall prompts provide richer and more accurate reports than direct and option-posing questions (Lamb et al., 2011; Malloy et al., 2011). Free recall prompts support free memory recall, enhance meta-cognitive mechanisms (cognitive processes involved in monitoring and regulating cognitive activities), and promote reporting. The collaborative nature of free recall prompts strengthens the child’s sense of security and willingness to disclose, fostering comfort in sharing sensitive details. This, in turn, improves the quality and accuracy of information disclosed during the interviews (Henderson, Russo, & Lyon, 2020). However, it is important to stress that studies have found direct questions to be effective, particularly among younger children (i.e., 3-6-year-olds; Hershkowitz et al., 2012). The reason relates to preschoolers requiring clues to aid them in retrieving memories of the CSA events. Therefore, direct questions by forensic interviewers may help them advance their report without compromising accuracy (Hershkowitz et al., 2012).

Based on previous studies, it appears that, alongside the effectiveness of free recall prompts and direct questions, the richness and accuracy of children's responses were found to gradually decrease with option-posing and suggestive questions (Hershkowitz et al., 2014; Lamb et al., 2011; Malloy & Quas, 2009). While option-posing and suggestive questions are discouraged due to their potential to elicit distorted information, follow-up questions can help mitigate these effects. It is crucial to handle suggestive questions with extreme care to uphold the integrity and reliability of information in forensic settings (Lamb et al., 2011; Malloy & Quas, 2009).

This study aimed to examine summary statements as a unique strategy to prompt detailed responses from children, alongside other question types, including free recall prompts. While Summaries are designed to elicit additional information with precision-such as the example, “Then Dad touched you under your clothes and then went out” in response to the child saying, “Dad lay me down on the bed and he touched me under my clothes and then went out.” the analysis encompassed a broader range of questioning strategies. Distinguishing between summaries and free recall prompts, allowed for the investigation of the unique impact of summaries on RR while also considering the effects of other question types. This comprehensive approach emphasizes the specific role of summaries, as well as the overall effectiveness of various questioning strategies in obtaining information from during forensic interviews.

The Current Study

The current study examined the relationship between child characteristics, abuse characteristics, social affiliation, and RR in forensic interviews. Hypotheses included a positive correlation between a child’s age and RR (H1), higher RR in girls’ reports versus boys (H2), and a link between higher abuse severity and lower RR (H3). Perpetrator’s identity will be associated with RR, with closer perpetrators yielding fewer details (H4). Abuse frequency will also influence RR, with richer reports from those who experienced a one-time abuse event compared to repeated events (H5). Social affiliation will play a role, with children from collectivist societies (e.g., Arab Muslim and ultra-Orthodox) expected to provide less detailed reports compared to those from individualistic societies (e.g., religious Jewish and secular Jewish) (H6). Finally, a general model will aim to test the relations between child characteristics, abuse characteristics, and social affiliation, on the one hand, and RR for all different question types: free recall prompts, summaries, directive, option-posing and suggestive. It is expected that RR will be lower if the preparator was the parent (compared to a stranger), among Arab-Muslims (compared to secular Jewish), and when the abuse included penetration compared to exposure without contact, in response to all question types.

Method

Study Sample

The sample included 314 forensic interview transcripts from 2015, conducted by 40 child forensic interviewers who were social workers trained in the NICHD protocol. These interviewers participated in training sessions led by certified instructors within the Ministry of Welfare specializing in child forensic interviews. The interviews took place either in the children’s schools for cases involving extrafamilial perpetrators or in the interviewer’s office for cases involving intrafamilial perpetrators.

The interviews met the following criteria: the children revealed they had been sexually abused; aged 3–13 at the time of the interview, in accordance with the ages of children interviewed in the Child Investigation Service; no evidence of developmental disturbances; and identified as either secular, religious, ultra-Orthodox Jewish, or Arab Muslim children. All children were interviewed in their mother tongue, with only the interviewer present. To ensures equal representation of transcripts from children in each of the social groups, “direct sampling” was employed. This method involved systematically selecting transcripts to reflect the diversity of the population under study, thereby enhancing the robustness and generalizability of the findings. In the final sample, there were 94 secular Jewish children (29.9%), 75 religious Jewish children (23.9%), 71 ultra-Orthodox Jewish children (22.6%) and 74 Arab Muslim children (23.6%).

Standardization of the Forensic Interviews

All forensic interviews followed the NICHD protocol, a standardized method for questioning child victims of CSA (Lamb et al., 2008). Studies have consistently shown that this protocol significantly improves the quality of forensic interviews compared to those conducted without it (Brown et al., 2019; Lamb et al., 2011; Leach et al., 2017). It aids investigators in evaluating the level of probability of abuse and enhances the credibility of the child’s report (Hershkowitz et al., 2007). Moreover, the protocol helps elicit investigative leads and contributes to case resolution (Cyr, 2022; Pipe et al., 2008).

Data Coding

To analyze the children’s narratives, the interviews were converted into data for statistical analysis. The content analysis categorized information into the following categories: (1) Age – noted in years. (2) Gender – male and female. (3) Social affiliation (self-reported) – Arab Muslim, secular Jew, religious Jew, ultra-Orthodox Jew. (4) Type of abuse – exposure, touching over clothing, touching under clothing, and penetration. (5) Identity of the perpetrator – parent, family member, acquaintance, and stranger.

The seventh variable, richness of reports, related to the number of details provided during the forensic interview. A detail is defined as information about the specific event the child is interviewed about (Lamb et al., 2011; Malloy et al., 2011). Each detail is marked only at its first appearance in the child’s narrative. The following is an example of how the details were marked: “Dad came into the room and took my shirt off.” In this example, the child provided six details: Dad, came in, room, took off, my, shirt.

In the present study, RR was examined based on question types used during the interviews: general richness of the reports (RR general; i.e., the total number of details the child provided in response to all types of questions asked), and richness of the reports in response to free recall prompts (RR free recall), summaries (RR summaries), direct questions (RR direct), option-posing questions (RR option-posing), and suggestive questions (RR suggestive). The eighth variable, question type, was encoded as follows: free recall prompts, summaries, direct questions, option-posing, and suggestive questions. It should be noted that all off-topic responses were excluded or disregarded from the analysis to ensure the focus on information related to the study.

Several steps were taken to estimate the reliability of the analysis. First, the two researchers who conducted the coding had formal NICHD training. Next, each transcript was systematically coded, and the reliability of this coding was examined by randomly re-coding one-fifth of the transcripts (two additional researchers, integral members of the larger research team, were involved in coding these transcripts to verify reliability). Disagreements regarding coding were resolved through discussion. The current study attained high interrater reliability using Cohen’s kappa scores (kappas > =.85) for coding categories.

Ethical Issues

This study received authorization from the Service for Forensic Interviews of Children, the Department of Welfare and Social Services and the Ethics Committee of the authors’ affiliated university. All identifying details were omitted from the transcripts to preserve the participants’ confidentiality before they were provided to the researchers conducting the study.

Methods of Data Analysis

Different analyses were used to test each hypothesis: Kruskal-Wallis tests and Bonferroni’s post-hoc analyses compared RR measures across abuse severities (Hypothesis 3). Similar analyses were used to compare RR measures across perpetrator identities (Hypothesis 4). Mann-Whitney U and Moses test of extreme reaction compared RR measures between children abused once and multiple times (Hypothesis 5). Kruskal-Wallis tests compared RR measures across social affiliations (Hypothesis 6). The general model was examined through a series of linear regression analyses predicting each RR measure (general, free recall prompts, direct, opting-posing, suggestive and summaries) according to child characteristics, abuse characteristics, and social affiliation. Initial Mann-Whitney U tests compared RR measures between girls and boys (Hypothesis 2). Age and gender (Hypotheses 1–2) were tested in the final regression model.

Results

The RR model revealed the following means and standard deviations for the research variables: Participants’ mean age was 9.5 years (SD = 2.54). The mean RR general was 114.88 details (SD = 76.11). Notably, the highest RR mean was in response to free recall prompts (M = 36.42, SD = 27.31), while the lowest was in response to suggestive questions (M = 0.64, SD = 1.69). The RR means for other question types were direct (M = 34.29, SD = 26.48), option-posing (M = 15.16, SD = 12.94), and summaries (M = 5.88, SD = 8.87).

Preliminary Analyses

From the preliminary analyses of the type of distribution, it appeared that the RR measures showed an asymmetrical distribution. Therefore, a series of a-parametric tests, Mann-Whitney U and Moses test of extreme reaction were conducted. Mann-Whitney U tests are the a-parametric alternative to t-tests for independent samples to compare two independent groups (i.e., boys, girls) in measures that are not normally distributed. Additionally, a Moses test of an extreme reaction allows for comparing two groups in anomalous cases. These cases represented forensic interviews that exhibited exceptionally high levels of detail compared to the average number of details typically provided in response to specific questions during interviews, specifically exceeding three standard deviations above the sample mean. Therefore, this analysis examined both the variations in RR and the frequency of interviews that displayed unusually high levels of detail in their responses.

H1. A Positive Relationship Will be Found Between Age and Richness of Reports

First, the distribution of RR measures was examined by a series of Kolmogorov-Smirnov tests to investigate the normal distribution with Lilliefors correction for the significance of the tests. The tests showed that the total RR measures did not have a normal distribution (Ds₍₃₁₄₎ > .11, ps < .001) but, rather, a positive asymmetric distribution. Thus, in most of the forensic interviews, RR was relatively low. Accordingly, to examine the relations between age and RR, a series of Spearman-type correlation analyses, suitable for variables that are not normally distributed, was conducted. The findings showed that, in accordance with H1, the older the abused child, the greater the RR summaries (r_s = .17, p = .002). However, in contradiction to H1, the older the abused child, the lower the RR direct (r_s = −.12, p = .037), RR option-posing (r_s = −.30, p < .0001), and RR suggestive (r_s = −.13, p = .02). Age was not found to be significantly related to RR general (r_s = −.09, p = .104) or RR free recall (r_s = .10, p = .096). Hence, this hypothesis was not supported.

H2. Richness of Girls’ Reports Will be Higher Than Boys

Contrary to the hypothesis, girls provided lower RR general and lower RR direct compared to boys (see Table 1). However, according to the hypothesis, the number of interviews with relatively high RR suggestive and RR summaries was higher among girls than boys. No significant differences between girls and boys were found in the other RR measures. Therefore, this hypothesis was partly supported.

Table 1.

Scores of Richness Measures According to Boys and Girls.

	Boys (n = 118)	Girls (n = 196)	Moses	Mann-Whitney U	Z
RR general	173.5	147.87	312.0	9676.0	−2.42*
RR free recall	151.40	167.64	309.0	10,367.5	−1.54
RR direct	174.75	147.12	313.0	9529.0	−2.61**
RR option-posing	164.40	153.35	311.0	10,750.0	−1.05
RR suggestive	153.28	160.04	192.***	12,062.0	0.84
RR summaries	160.29	155.82	284.0*	11,234.5	−0.43

Note. The measures reported in the columns of boys and girls are the scores of mean measures. *p < .05. **p < .01. ***p < .001. Z = Mann-Whitney test value articles for Z distribution. RR = richness of reports.

H3. Richness of Reports Will Differ Depending on Severity of Abuse

A series of Kruskal-Wallis a-parametric tests were conducted, which constitute an a-parametric alternative to one-way analysis of variance (ANOVA) in light of the asymmetry of the RR measures. The results are presented in Table 2. The findings showed significant differences between RR general, RR free recall and RR direct and cases with different abuse severities. Bonferroni’s post-hoc analyses showed that RR general and RR direct were significantly higher when penetration occurred compared to contactless exposure, contact over clothing and contact under clothing. In addition, RR free recall was higher when penetration occurred compared to contactless exposure. In the remaining RR measures, no significant differences were found. Therefore, this hypothesis was not supported.

Table 2.

Scores of Richness According to Type of Abuse With Test Statistics and Post-hoc Analyses.

	Contactless exposure (n = 62)	Contact over clothes (n = 51)	Contact under clothes (n = 151)	Penetration (n = 45)	χ2 (3)
RR general	138.10^b	136.91^b	153.46^b	203.97^a	17.87**
RR free recall	140.14^b	144.23	154.01	191.01^a	9.79*
RR direct	132.28^b	141.01^b	153.75^b	206.17^a	20.01***
RR option-posing	153.20	150.47	161.97	161.46	0.89
RR suggestive	148.59	154.37	159.57	161.11	1.36
RR summaries	153.98	155.17	154.23	174.90	2.01

Note. The measures reported in the columns of the type of abuse are scores of mean measures. *p < .05. **p < .01. ***p < .001. Means that significantly differ from each other (p < .05) are marked with different letters. RR = richness of report.

H4. Richness of Reports Will Differ Depending on Perpetrator’s Identity

A series of Kruskal-Wallis a-parametric tests were conducted. The results are presented in Table 3. The findings indicated differences in RR general and RR free recall in relation to the different perpetrator identities. Bonferroni’s post-hoc analyses showed, in accordance with the hypothesis, RR general and RR free recall were significantly lower when the perpetrator was the parent compared to an acquaintance who was not a family member. In the other RR measures, no significant differences were found.

Table 3.

Scores of Richness According to Identity of the Perpetrator With Test Statistics and Post-hoc Analyses.

	Parent (n = 37)	Family member (n = 65)	Acquaintance (n = 110)	Stranger (n = 101)	χ² (3)
RR general	119.24^a	161.79	171.09^b	155.35	9.27*
RR free recall	117.85^a	158.05	173.20^b	155.63	10.38*
RR direct	125.22	163.45	166.26	157.49	6.01
RR option-posing	155.91	162.17	154.70	159.64	0.34
RR suggestive	148.45	160.20	157.85	159.10	0.78
RR summaries	127.55	154.38	159.67	169.36	5.99

Note. The measures reported in the columns of the perpetrator’s identity are the average scores of the measures. *p < .05. Means that significantly differed from each other (p < .05) are marked with different letters. RR = richness of report.

H5. Richness of Reports of Children Abused Once Will Be Higher Than Children Abused Multiple Times

A series of a-parametric tests were conducted, including the Mann-Whitney U and Moses test of extreme reaction. The results are presented in Table 4. No significant differences were found in the scores of the different RR measures between one-time and multiple abuse events. While the data indicated that RR was somewhat higher in cases of one-time abuse across several measures-namely, RR general, RR free recall, RR option-posing, RR suggestive and RR summaries-the differences were minimal and did not reach statistical significance. Therefore, this hypothesis was not supported, and the observed variations should be interpreted with caution.

Table 4.

Scores of Richness According to Frequency of Abuse.

	One-time abuse (n = 161)	Multiple abuse events (n = 152)	Moses	Mann-Whitney U	Z
RR general	157.25	156.73	264.0**	12,195.0	−0.05
RR free recall	159.72	154.12	261.0**	11,797.5	−0.55
RR direct	157.40	156.58	275.0	12,172.0	−0.08
RR option-posing	147.55	167.01	269.0*	13,757.0	1.90
RR suggestive	158.93	154.96	184.0***	11,926.0	−0.51
RR summaries	160.32	153.48	263.0**	11,701.0	−0.67

Note. The measures reported in the columns of frequency of abuse are the average scores of the measures. ***p < .001, **p < .01*p < .05. RR = richness of report.

H6. Richness of Reports from Arab Muslim and Ultra-Orthodox Jewish Children Will Be Lower Than Secular and Religious Jewish Children

A series of a-parametric Kruskal-Wallis tests were conducted. The results are presented in Table 5. RR general, RR free recall, RR direct and RR option-posing were found to be higher for Arab Muslim children than for children of all Jewish origins (secular, religious and ultra-Orthodox). No significant differences were found in RR option-posing and RR summaries. Therefore, this hypothesis was not supported.

Table 5.

Scores of Richness According to Social Affiliation With Test Statistics and Post-hoc Analyses.

	Arab Muslim (n = 74)	Secular Jewish (n = 94)	Religious Jewish (n = 75)	Ultra-Orthodox Jewish (n = 71)	χ ² ₍₃₎
RR general	214.84^a	133.38^b	150.45^b	137.11^b	40.19***
RR free recall	205.76^a	151.44^b	146.56^b	126.78^b	30.56***
RR direct	199.69^a	136.54^b	146.56^b	146.48^b	22.26***
RR option-posing	203.74^a	127.46^b	149.60^b	157.42^b	30.10***
RR suggestive	167.56	156.76	142.37	163.98	5.75
RR summaries	157.17	150.86	169.27	154.20	1.88

Note. The measures reported in the columns of social affiliation are the scores of mean measures. ***p < .001. Means that significantly differed from each other (p < .05) are marked with different letters. RR = richness of report.

Summarizing Model to Examine the Explained Variance in Richness of Reports

A series of multiple regression analyses were performed to test whether the study’s variables were associated with the subtypes of RR. The regression coefficients and degree of significance for the regressions are presented in Table 6.

Table 6.

Standardized Regression Coefficients (Beta) (With CI 95%) and Degree of Significance for Richness With all Variables.

RR variables	General	Free recall prompts	Direct questions	Option-posing questions	Suggestive questions	Summaries
Age	−.08 (−0.20, 0.04) SE = .02	.07 (0.01, 0.13), SE = .02	−.13 (−0.26, −0.01), SE = .01	−.21** (−0.35, −0.07), SE = .08	−.14 (−0.29, 0.02), SE = .07	.15 (0.09, 0.21), SE = .07
Gender	−.02 (−0.11, 0.07) SE = .03	.04 (−0.02, 0.10), SE = .01	−.06 (−0.16, 0.03), SE = .03	−.01 (−0.11, 0.09), SE = .01	.09 (−0.02, 0.20), SE = .05	.01 (−0.07, 0.09), SE = .01
Identity of the preparator
Parent versus relative	.22** (0.08, 0.36), SE = .01	.20* (0.05, 0.35), SE = .05	.20* (0.05, 0.35), SE = .05	.16 (0.01, 0.31), SE = .07	.16 (0.01, 0.31), SE = .05	.05 (−0.08, 0.18), SE = .02
Parent versus acquaintance	.26** (0.12, 0.40), SE = .05	.26** (0.11, 0.41), SE = .06	.24** (0.09, 0.39), SE = .06	.11 (−0.04, 0.26), SE = .05	.17 (0.02, 0.32), SE = .05	.05 (−0.08, 0.18), SE = .02
Parent versus stranger	.29*** (0.15, 0.43), SE = .06	.19* (0.04, 0.34), SE = .04	.31*** (0.16, 0.46), SE = .07	.28*** (0.13, 0.43), SE = .08	.24** (0.09, 0.39), SE = .06	.23** (0.08, 0.38), SE = .07
Social affiliation
Secular Jew versus Arab Muslim	.38*** (0.24, 0.52), SE = .05	.24** (0.09, 0.39), SE = .06	.28***(0.13, 0.43), SE = .08	.41*** (0.26, 0.56), SE = .07	.16 (0.01, 0.31), SE = .07	.07 (−0.07, 0.21), SE = .05
Secular Jew versus religious Jew	−.01 (−0.15, 0.13), SE = .02	−.09 (−0.23, 0.05), SE = .03	−.04 (−0.19, 0.10), SE = .01	.05 (−0.09, 0.19), SE = .03	−.04 (−0.18, 0.10), SE = .01	.10 (−0.04, 0.24), SE = .06
Secular Jew versus ultra-Orthodox Jew	−.01 (−0.15, 0.13), SE = .01	−.11 (−0.25, 0.03), SE = .01	.00 (−0.15, 0.14), SE = .00	.07 (−0.07, 0.21), SE = .03	.04 (−0.10, 0.18), SE = .01	.10 (−0.04, 0.24), SE = .05
Type of abuse
Contactless exposure versus contact over clothes	−.03 (−0.16, 0.09), SE = .01	−.03 (−0.16, 0.09), SE = .01	−.02 (−0.15, 0.11), SE = .01	.00 (−0.13, 0.13), SE = .00	.00 (−0.13, 0.13), SE = .00	−.02 (−0.15, 0.11), SE = .01
Contactless exposure versus contact under clothes	.10 (−0.04, 0.24), SE = .04	.07 (−0.07, 0.21), SE = .01	.11 (−0.03, 0.25), SE = .01	.12 (−0.02, 0.26), SE = .04	.08 (−0.06, 0.22), SE = .02	.06 (−0.08, 0.20), SE = .02
Contactless exposure versus penetration	.23** (0.08, 0.38), SE = .06	.19* (0.04, 0.34), SE = .04	.25**(0.10, 0.40) SE = .06	.01 (−0.14, 0.16), SE = .01	.07 (−0.08, 0.22), SE = .03	.17* (0.02, 0.32), SE = .04
R2	24.0%	16.7%	20%	22.3%	6.8%	8.0%

Note. *p < .05, **p < .01, ***p < .001. RR = richness of report.

RR general

In the model predicting RR general, the general model was significant, F (10, 228) = 12.25, p < .001, and the explained variance was 24%. When the parent was the perpetrator compared to a relative, acquaintance, or stranger, RR general was lower. The RR was higher when the abuse included penetration compared to exposure without contact. The RR did not differ between contactless exposure and contact under or over clothes. Additionally, it was found that the RR of the Arab Muslim children was higher compared to secular Jews. No significant differences were found between secular Jews and the other ethnoreligious groups. Age, gender, and type of abuse were not significant.

RR Free Recall Prompts

In the model predicting RR free recall, the general model was significant, F (10, 228) = 9.5, p < .001, and the total explained variance was 16.7%. When the parent was the perpetrator compared to a relative, acquaintance, or stranger, the RR was lower. The RR was higher when the abuse included penetration in contrast to exposure without contact. The RR did not differ between contactless exposure and contact under or over clothes. Additionally, it was found that the RR of the Arab Muslim children was higher compared to secular Jews. No significant differences were found between secular Jews and the other ethnoreligious groups. Age, gender, and type of abuse were not significant.

RR Direct Questions

In the model predicting RR direct, the general model was significant, F (10, 228) = 11.5, p < .001, and the total explained variance was 20%. When the parent was the perpetrator compared to a relative, acquaintance, or stranger, the RR was lower. The RR was higher when the abuse included penetration in contrast to exposure without contact. However, the RR did not differ between contactless exposure and contact under or over clothes. Additionally, it was found that the RR of the Arab Muslim children was higher compared to secular Jews. No significant differences were found between secular Jews and the other ethnoreligious groups. Age and gender were not significant.

RR option-posing questions

In the model predicting RR option-posing, the general model was significant, F (10, 228) = 13.22, p < .001, and the total explained variance was 22.3%. When the parent was the perpetrator compared to a stranger, the RR was lower. No differences were found when the preparator was a parent compared to a relative or acquaintance. Additionally, it was found that the RR of the Arab Muslim children was higher compared to secular Jews. No significant differences were found between secular Jews and the other ethnoreligious groups. The type of abuse was not associated with the RR. Older age was associated with significantly lower RR, while gender was not significant.

RR Suggestive Questions

In the model predicting RR suggestive, the general model was significant, F (10, 228) = 2.57, p = .006, and the total explained variance was 6.8%. The only significant factor was the identity of the preparator. Specifically, when the parent was the perpetrator compared to a stranger, the RR was lower. No differences were found when the preparator was a parent compared to a relative or acquaintance. Additionally, social affiliation, type of abuse, gender, and age were not associated with the RR.

RR Summaries

In the model predicting RR summaries, the general model was significant, F (10, 228) = 2.14, p = .003, and the total explained variance was 8%. The identity of the perpetrator was significant. Specifically, when the parent was the perpetrator compared to a stranger, the RR was lower. No differences were found when the preparator was a parent compared to a relative or acquaintance. The RR was higher when the abuse included penetration in contrast to exposure without contact. The RR did not differ between contactless exposure and contact under or over clothes. Social affiliation, gender, and age were not associated with the RR.

Discussion

The current study aimed to examine the ways the richness of children’s reports in cases of CSA was impacted by their social affiliation. Therefore, the relationship between the social affiliation of sexually abused children and the richness of their reports in forensic interviews was explored. In addition, the children’s characteristics and abuse characteristics were also examined in relation to the richness of their reports.

Social Affiliation and Richness of Reports

The current findings revealed higher general RR among Arab Muslim children compared to their Jewish counterparts, contrary to the established literature. These unexpected results underscore the intricate interplay between cultural affiliations, societal norms, and religious influences on forensic interviews. Expanding this analysis to a broader global context, similar dynamics can be observed in countries with multicultural populations (Ettinger, 2024). In the U.S., for instance, the richness of the reports provided by children from different cultural backgrounds has been found to be influenced by a variety of factors. Children from Asian American communities, for example, may exhibit communication styles and disclosure patterns that differ from children from African American or Native American backgrounds or other groups, such as the Amish (McGuigan & Stephenson, 2015; Rapholo & Makhubele, 2019). Cultural values, such as respect for authority, family hierarchy, and communal harmony, can impact how children engage in forensic interviews and provide information. In countries like Canada, Australia, and the U.K., where multiculturalism is a defining feature, the RR in forensic interviews can vary based on cultural affiliations. Studies have shown that children from Indigenous communities, immigrant families, or marginalized ethnic groups may have unique communication styles and preferences for disclosing sensitive information (Dando et al., 2023; Deshmukh, 2019). Understanding the cultural nuances and societal norms of diverse populations is crucial for conducting effective and culturally sensitive forensic interviews.

Furthermore, regarding the moderating factor of the question type, in contrast to the hypothesis, the RR in response to free recall prompts and direct and option-posing questions among the Arab Muslim children was higher than that of the Jewish children, with no differences among the Jewish groups. Previous studies have found that children from more collectivist cultures exhibit less detailed reports than those from more individualistic cultures (e.g., Hindi et al., 2022). Potential explanations for this unexpected outcome center around the societal values among Arab Muslim children. Traditional Arab societal norms emphasize obedience, discipline, and deference to adult authority (Elias & Sulimani-Aidan, 2024). This cultural framework may influence how these children engage in forensic interviews, aligning with familial and societal expectations, as supported by recent qualitative research (Katz et al., 2020). Another possible explanation relates to the significant role of religion in Muslim lives. Islamic law, known as Sharia, heavily influences decision-making based on the Quran and the Oral Law (Sunnah) (Mahajne et al., 2021). Sharia and the Quran strictly forbid perjury. Consequently, Arab Muslim children might be more inclined to cooperate during forensic interviews to avoid falsehoods, thereby aligning with religious teachings.

The absence of significant differences among the Jewish social groups was unexpected. This complexity might stem from the intricate nature of Jewish Israeli society. There is a lack of clear-cut distinction between secular, religious, and ultra-Orthodox communities (Shoham, 2014), each comprising diverse subgroups. Thus, personal identification along the religious spectrum might be complex and inconsistent (Shoham, 2014). This finding potentially suggests that forensic interviewers in Israel adeptly adapt the interview approaches to align with the cultural norms of the children.

By considering the impact of social affiliation on the RR in forensic interviews, researchers and practitioners can enhance their understanding of how cultural diversity influences children’s responses. Tailoring interview approaches to accommodate the cultural and social backgrounds and values of children from diverse communities can improve the quality and reliability of information obtained during forensic interviews. This inclusive approach promotes fairness, accuracy, and cultural competence in the forensic interview process on a global scale. Additionally, integrating considerations specific to social affiliations into established protocols, like the NICHD forensic protocol standards, can further enhance the effectiveness and cultural sensitivity, ensuring that children from various cultural backgrounds receive appropriate support and are able to provide detailed and accurate reports.

Characteristics of the Child and Richness of Reports

The current study did not reveal a significant relation between a child’s age and RR, contrary to previous literature (Hershkowitz et al., 2012, 2014; Lamb et al., 2011). This may be explained in terms of developmental differences influenced by genetics, environment, and personal experiences (e.g., Daniel, 2014). These variations in cognitive and linguistic development can affect information recall during interviews, potentially influencing the quality of reports. Therefore, the complex developmental differences among children may explain the absence of a definitive link between age and RR.

In the moderation model, the age of the abused child did not significantly correlate with RR in free recall prompts. Yet, older abused children provided richer reports in response to summaries, supporting the hypothesis (H1). This highlights the importance of summaries in eliciting detailed narratives, as their precision can trigger retrieval cues and enhance the content. The study aligns with Tulving’s (1983) context-dependent memory theory, emphasizing the retrieval cue’s role in forensic interviews with abused children. It also confirms that younger children provide richer reports in response to direct, option-posing, and suggestive questions, consistent with prior research (e.g., Hershkowitz et al., 2012). Younger children benefit from specific requests in direct questions as memory cues, aiding recall compared to free recall prompts and reducing suggestiveness. Minimizing option-posing and suggestive questions is crucial to promote accurate reporting in younger children, considering their limited meta-cognitive abilities (Lamb et al., 2011; Malloy & Quas, 2009).

The study did not find gender differences in RR, with girls providing less detailed reports than boys, contrary to previous research (Cross et al., 2021; Leach et al., 2017). This outcome may be explained societally, as studies have suggested that boys receive more maternal support after sexual abuse, positively impacting their reporting (Hershkowitz et al., 2014; Pintello & Zuravin, 2001). Maternal support can alleviate negative effects like guilt, anxiety, and stress, encouraging children to cooperate with interviewers and disclose more abuse details (Hershkowitz et al., 2014; Lamb et al., 2008).

Characteristics of the Abuse and Richness of Reports

The study partially confirmed that RR differs based on the perpetrator’s identity, with lower general RR observed when the perpetrator was a parent compared to a non-family member, in line with prior research (Leach et al., 2017; Lippert et al., 2009). The betrayal theory highlights the child’s loyalty and conflict towards the perpetrator, intensifying the struggle to report despite the abuse severity, and impacting cooperation in forensic interviews (Katz & Barnetz, 2014; Katz et al., 2020; Lawson & Akay-Sullivan, 2020).

Examination of the moderating effect of question types revealed a significant decrease in RR in response to free recall prompts when the perpetrator was a parent versus an acquaintance, aligning with the hypothesis. This finding is unsurprising as free recall prompts require cooperative motivation and can trigger loyalty conflicts in children between their parent and themselves, leading to less detailed reports against the abusive parent (Hershkowitz et al., 2005; Lamb et al., 2011; Malloy et al., 2011).

When examining the relation between abuse severity and RR, the hypothesis was not supported. It found that general RR significantly differed based on abuse severity, with notably higher richness in cases involving penetration compared to other forms of abuse. This finding aligns with prior research (Olafson, 2011) and suggests that cases involving penetration may be inherently more complex, leading to a greater number of details reported (Olafson, 2011). Penetrative sexual abuse is identified as the most traumatic form, positively impacting memory, potentially driven by adaptive memory functions that aid in avoiding future trauma (Olafson, 2011; Pipe et al., 2004). This type of abuse is associated with heightened psychological and social harm to survivors, inducing increased fear, insecurity, and helplessness in abused children (Gries et al., 2000; Li, 2019). In such a situation, children may cooperate with forensic interviewers and provide more detailed descriptions of the abuse due to feeling unable to cope with the associated challenging emotions and sensations, as they lack the adaptive mechanisms for survival.

The study’s results contradicted the hypothesis regarding question types, showing that RR in response to direct questions was significantly higher in cases involving penetration compared to other forms of abuse. Similarly, richer reports were observed with free recall prompts when penetration occurred. These findings support the notion that children may provide more detailed accounts in cases of penetrative abuse due to the overwhelming nature of the experience. Previous literature suggests that in severe abuse situations, children may demonstrate improved memory recall abilities, with forensic interviewers’ use of free recall prompts and direct questions enhancing information retrieval (Korkman et al., 2006).

The hypothesis that children who experienced a single abuse incident would provide richer reports compared to those who experienced multiple events was not supported. Surprisingly, no significant differences were found in RR between one-time and multiple abuse events, nor in response to question type, which contrasts with previous research (e.g., Burbacher & La Rooy, 2014). While we measured total details reported, it is plausible that a child who experienced a single incident might report more specifics about that event. Conversely, a child with multiple incidents may provide a greater total number of details. The lack of significant differences may also be influenced by factors such as the perpetrator’s identity, abuse type, and the child’s social affiliation. Further research is needed to explore the relationship between abuse frequency and RR, considering the specificity of details reported.

Limitations and Recommendations

While the current study innovatively explored the richness of reports in forensic interviews with sexually abused children concerning abuse characteristics and social affiliation, its limitations warrant acknowledgment. First, relying on forensic interviews, unlike controlled laboratory studies, lacks the ability to cross-verify the information provided by the child. Second, categorizing Jewish children into religious groups based on self-reports may overlook the complexity of individual identities, potentially accounting for the subtle differences observed between religious and secular Jews in this study.

A third limitation relates to the highly diverse Arab Muslim population in Israel (Fahoum & Al-Krenawi, 2023). This diversity was not captured in this study, as the Arab Muslim participants were included only in a single category. For example, many Bedouin Arabs are still semi-nomadic, and there are significant differences between Muslims living in primarily Arab municipalities, religiously mixed Arab municipalities, and predominantly Jewish municipalities in Israel (Ghara, 2013).

Finally, despite efforts to reduce intervening variables, unexplored factors might have affected the children’s decisions to provide detailed reports. Variables like meta-cognitive abilities, perpetrator age, parental support, the gender of the child forensic interviewer, and the time lapse between the abuse and report were not analyzed but could play roles. Future studies should investigate these aspects.

Theoretical and Practical Contributions

The current study offers theoretical and practical contributions to the field of CSA. At the theoretical level, the findings contribute to a deeper understanding of the way social characteristics are likely to influence how child survivors of sexual abuse function in the framework of forensic interviews. These findings strengthen the importance of continuing context-aware research that promotes a holistic understanding.

In addition, the various contexts examined in this study highlight the importance of acknowledging intersectionality and the need for intersectional investigations in future studies. There is great meaning and consequence in understanding the phenomenon of CSA and the experiences of sexually abused children alongside their multiple environments of marginalization, such as ethnic, religious, social and gender. A conscious investigation of their contexts and intersectionalities of marginalization, along with analyses of their relations, would likely promote an improved and more profound understanding of CSA.

At the practical level, the findings attested to the contribution of the different types of questions to abused children from different social affiliations in providing detailed reports. A context-aware observation of CSA, as shown in this study, could promote a better understanding of the phenomenon and lead to the development of services adapted to children of different cultures. In addition, the findings regarding the contribution of the different question types have the potential to advance the practice of various professionals, not only in obtaining reports and ensuring justice, but also in clinical interventions.

In conclusion, this groundbreaking study explored how a child’s social affiliation impacts the richness of reports in forensic interviews of sexually abused children, highlighting the multifaceted nature of reporting influenced by various factors. Emphasizing the importance of context and intersectionality, the study underscores the necessity of considering these aspects to enhance understanding of child sexual abuse (CSA) and to tailor interventions and services effectively.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Hanin Mordi

References

Alaggia

Collin-Vézina

Lateef

(2019). Facilitators and barriers to child sexual abuse (CSA) disclosures: A research update (2000–2016). Trauma, Violence & Abuse, 20(2), 260–283. https://doi.org/10.1177/1524838017697312.

Al-amer

Dwekat

Ali

Abuzied

Alzahrani

N. S.

Alhowaymel

F. M.

Alharbi

H. F.

Lapadula

AlBashtawy

Hussein

M. M.

Randall

(2024). Prevalence of stress and types of coping strategies among adolescents (14–18 years) in collectivist communities. Journal of Pediatric Nursing, 1(1), 2.

Ben-Porat

Feniger

Filc

Kabalo

Mirsky

(Eds.), (2022). Routledge handbook on contemporary Israel. Routledge, Taylor & Francis Group.

Brown

D. A.

Lewis

C. N.

Lamb

M. E.

Gwynne

Kitto

Stairmand

(2019). Developmental differences in children’s learning and use of forensic ground rules during an interview about an experienced event. Developmental Psychology, 55(8), 1626–1639. https://doi.org/10.1037/dev0000756.

Burbacher

S. P.

La Rooy

(2014). Witness recall across repeated interviews in a case of repeated abuse. Child Abuse & Neglect, 38(2), 202–211.

Cahaner

(2023). The study of Haredi space in Israel: Trends, characteristics, achievements, and challenges. In Caplan

Leon

(Eds.), Contemporary Israeli Haredi Society (pp. 125–159). Routledge. https://doi.org/10.4324/9781003315643

Central Bureau of Statistics . (2021). Shnaton statisti leIsrael (Gilion mispar 72) [Statistical yearbook for Israel. Central Bureau of Statistics. Report No. 72)].

Cross

T. P.

Ernberg

Walsh

W. A.

(2021). The criminal justice response to child and youth victimization. In Geffner

White

J. W.

Hamberger

L. K.

Rosenbaum

Vaughan-Eden

Vieth

V. I.

(Eds.) Handbook of interpersonal violence and abuse across the lifespan, (pp. 1797–1827). Springer.

Cyr

(2022). Conducting interviews with child victims of abuse and witnesses of crime: A practical guide (1st ed.). Routledge. https://doi.org/10.4324/9781003265351.

10.

Dando

C. J.

Taylor

P. J.

Sandham

A. L.

(2023). Cross cultural verbal cues to deception: Truth and lies in first and second language forensic interviews. Frontiers in Psychology, 14(1), 1152904. https://doi.org/10.3389/fpsyg.2023.1152904

11.

Daniel

(2014). Hitpatchut hayeled. Century - Department of Pedagogical and Technological Development. [Child Development].

12.

David

Khatib

(2021). Differences between Israeli Jewish and Arabs mothers in parental self-efficacy: A cultural perspective. Social Work and Social Welfare, 3(1), 121–133. https://doi.org/10.25082/SWSW.2021.01.005.

13.

Deshmukh

R. R.

(2019). What are children telling us? A cross cultural study of children's suggestibility [Unpublished master's thesis]. Cornell University.

14.

Easton

S. D.

Saltzman

L. Y.

Willis

D. G.

(2014). “Would you tell under circumstances like that?”: Barriers to disclosure of child sexual abuse for men. Psychology of Men & Masculinity, 15(4), 460–469. https://doi.org/10.1037/a0034223.

15.

Elias

Sulimani-Aidan

(2024). “Adopting unsuitable solutions constitutes an additional risk factor”: Context-informed social work practice with at-risk young adult Arabs. Children and Youth Services Review, 161(1), 107616.

16.

Ettinger

(2024). Enhancing cultural safety in forensic interviews during investigations of child abuse : Indigenous perspectives. University of British Columbia. [Unpublished doctoral dissertation].

17.

Fahoum

Al-Krenawi

(2023). Perceptions of stigma toward mental illness in Arab society in Israel. International Social Work, 66(3), 769–780.

18.

Freyd

J. J.

(1996). Betrayal trauma. Harvard University Press.

19.

Gemara

Nadan

(2022). “He who spareth the rod hateth his son”: Perceptions regarding corporal punishment among ultra-Orthodox Jewish fathers in Israel. Journal of Interpersonal Violence, 37(1-2), 79–101.

20.

Ghara

(2013). Book of Arab society in Israel (6): Population, society, economy. Van Leer Jerusalem Institute & Hakibbutz Hameuchad.

21.

Gerdes

K. E.

Beck

M. N.

Cowan-Hancock

Wilkinson-Sparks

(1996). Adult survivors of childhood sexual abuse: The case of Mormon women. Affilia, 11(1), 39–60. https://doi.org/10.1177/088610999601100104.

22.

Gries

L. T.

Goh

D. S.

Andrews

M. B.

Gilbert

Praver

Naierman-Stelzer

(2000). Positive reaction to disclosure and recovery from child sexual abuse. Journal of Child Sexual Abuse, 9(1), 29–51. https://doi.org/10.1300/J070v09n01_03.

23.

Henderson

H. M.

Russo

Lyon

T. D.

(2020). Forensic interviewers’ difficulty with free recall prompts: Faux free recall prompts and negative recasting. Child Maltreatment, 25(3), 363–372. https://doi.org/10.1177/1077559519895593.

24.

Hershkowitz

Fisher

Lamb

M. E.

Horowitz

(2007). Improving credibility assessment of child sexual abuse allegation: The role of the NICHD investigating interview protocol. Child Abuse & Neglect, 31(2), 99–110.

25.

Hershkowitz

Horowitz

Lamb

M. E.

(2005). Trends in children’s disclosure of abuse in Israel: A national study. Child Abuse & Neglect, 29, 1203–1214.

26.

Hershkowitz

Lamb

M. E.

Katz

(2014). Allegation rates in forensic child abuse investigations: Comparing the revised and standard NICHD protocols. Psychology, Public Policy, and Law, 20(3), 336–344.

27.

Hershkowitz

Lamb

M. E.

Orbach

Katz

Horowitz

(2012). The development of communicative and narrative skills among preschoolers: Lessons from forensic interviews about child abuse. Child Development, 83(2), 611–622.

28.

Hindi

Mordi

Tener

Katz

(2022). How does religiosity affect the richness of child forensic testimonies? Comparing the narratives of sexual abuse victims from three Jewish groups in Israel. Children and Youth Services Review, 137, 106488.

29.

Jaraisi

(2012). Psychosocial treatment of Arab society. In Hovav

Levintal

Katan

(Eds.) Social work in Israel, (pp. 527–559). Hakibbutz Hameuchad & Association for the Advancement of Social Work.

30.

Katz

Barnetz

(2014). The behavior patterns of abused children as described in their testimonies. Child Abuse & Neglect, 38(6), 1033–1040.

31.

Katz

Tener

Marmor

Lusky-Weisrose

Mordi

(2020). Yes, my uncle, I'll do whatever you say”: Experiences of Israeli Muslim Arab children during forensic interviews following child sexual abuse. Journal of Interpersonal Violence, 37(5-6), NP2465–NP2489. https://doi.org/10.1177/0886260520943732.

32.

Keidar

Snir

Regev

Keidar

(2022). Ultra-Orthodox parents' perceptions of arts therapies for their children. Children, 9(10), 1576.

33.

Khoury

(2013). The transition to the independent life of trainees in religious-national boarding schools in Israel. The Hebrew University. [Unpublished master's thesis].

34.

Korkman

Santtila

Sandnabba

(2006). Dynamics of verbal interaction between interviewer and child in interviews with alleged victims of child sexual abuse. Scandinavian Journal of Psychology, 47(2), 109–120.

35.

Korkman

Juusola

Santtila

(2014). Who made the disclosure? Recorded discussions between children and caretakers suspecting child abuse. Psychology, Crime & Law, 20(10), 994–1004. https://doi.org/10.1080/1068316X.2014.902455.

36.

Lamb

M. E.

Hershkowitz

Orbach

Esplin

(2008). Tell me what happened: Structured investigative interviews of child victims and witnesses. Wiley.

37.

Lamb

M. E.

La Rooy

Malloy

L. C.

Katz

(2011). Children's testimony: A handbook of psychological research and forensic practice (2nd ed.). Wiley.

38.

Lawson

D. M.

Akay-Sullivan

(2020). Considerations of dissociation, betrayal trauma, and complex trauma in the treatment of incest. Journal of Child Sexual Abuse, 29(6), 677–696. https://doi.org/10.1080/10538712.2020.1751369.

39.

Leach

Powell

M. B.

Sharman

S. J.

Anglim

(2017). The relationship between children’s age and disclosures of sexual abuse during forensic interviews. Child Maltreatment, 22(1), 79–88. https://doi.org/10.1177/1077559516675723.

40.

(2019). Perceptual mechanisms of anxiety and its disorders. In Olatunji

(Ed.) Cambridge handbook of anxiety and related disorders, (pp. 59–87).

41.

Lippert

Cross

T. P.

Jones

L. M.

Walsh

(2009). Telling interviewers about sexual abuse: Predictors of child disclosure at forensic interviews. Child Maltreatment, 14(1), 100–113. https://doi.org/10.1177/1077559508318398.

42.

Mahajne

Alhuzail

N. A.

Bar-on

(2021). Arab Muslim social workers between the Sharia and Western interventions and their coping mechanisms. Journal of Religion & Spirituality in Social Work: Social Thought, 40(4), 459–474.

43.

Malloy

L. C.

Brubacher

S. P.

Lamb

M. E.

(2011). Expected consequences of disclosure revealed in investigative interviews with suspected victims of child sexual abuse. Applied Developmental Science, 15(1), 8–19.

44.

Malloy

L. C.

Quas

J. A.

(2009). Children’s suggestibility: Areas of consensus and controversy. In Kuehnle

Connell

(Eds.) The evaluation of child sexual abuse allegations: A comprehensive guide to assessment and testimony, (pp. 167–297). John Wiley & Sons.

45.

McGuigan

W. M.

Stephenson

S. J.

(2015). A single-case study of resiliency after extreme incest in an Old-Order Amish family. Journal of Child Sexual Abuse, 24(5), 526–537.

46.

Mehrotra

(2010). Toward a continuum of intersectionality theorizing for feminist social work scholarship. Affilia, 25(4), 417–430. https://doi.org/10.1177/0886109910384190

47.

Mordi

Katz

Tener

Savaya

(2022). Disclosing the abuse: The effect of ethnoreligious identity on CSA disclosure in forensic interviews. Child Abuse & Neglect, 124(2), 105441. https://doi.org/10.1016/j.chiabu.2021.105441

48.

Nadan

Ganz

(2018). Perspectives of ultra-Orthodox children in Israel on risk and protection: The intersection of culture, religion, spirituality, and socio-political context. Childhood, 25(3), 325–339. https://doi.org/10.1177/0907568218774096.

49.

Nadan

Gemara

Keesing

Bamberger

Roer-Strier

Korbin

(2019). ‘Spiritual risk’: A parental perception of risk for children in the ultra-Orthodox Jewish community. The British Journal of Social Work, 49(5), 1198–1215.

50.

Nadan

Spilsbury

J. C.

Korbin

J. E.

(2015). Culture and context in understanding child maltreatment: Contributions of intersectionality and neighborhood-based research. Child Abuse & Neglect, 41, 40–48. https://doi.org/10.1016/j.chiabu.2014.10.021

51.

Nijim-Ektelat

BenRabbi

Sabu-Lal

(2018). Ikronot avodah vehitarvut hamotamim lechevra haAravit beshirutei revacha vetipul beIsrael [Working and intervention principles adapted to the Arab society in welfare and treatment services in Israel]. Myers-Joint-Brookdale Institute & Engelberg Center for Children and Youth. https://brookdale.jdc.org.il/wp-content/uploads/2018/06/778-18-Heb-WEB-Report-Intervention-Practices-Adapted.pdf.

52.

Olafson

(2011). Child sexual abuse: Demography, impact and interventions. Journal of Child & Adolescent Trauma, 4(1), 8–21.

53.

Parkinson

(2014). Child sexual abuse and the churches: A story of moral failure? Current Issues in Criminal Justice, 26(1), 119–138.

54.

Piaget

(1936). Origins of intelligence in the child. Routledge & Kegan Paul.

55.

Pintello

Zuravin

(2001). Intrafamilial child sexual abuse: Predictors of post disclosure maternal belief and protective action. Child Maltreatment, 6, 344–352.

56.

Pipe

M. E.

Lamb

M. E.

Orbach

Esplin

P. W.

(2004). Recent research on children's testimony about experienced and witnessed events. Developmental Review, 24(4), 440–468. https://doi.org/10.1016/j.dr.2004.08.006.

57.

Pipe

M. E.

Orbach

Lamb

M. E.

(2008). The effect of the NICHD Interview Protocol on the outcomes of child sexual abuse investigations [Conference presentation]. American Psychology Law Society Conference.

58.

Poniscjakova

(2021). Challenges with opening up closed off communities: Interviewing ultra-Orthodox Jewish communities in Israel. In Islam

M. R.

Khan

N. A.

S. H. A. B.

Wahab

H. A.

Hamidi

M. B.

(Eds.) Field guide for research in community settings, (pp. 25–33). Edward Elgar Publishing.

59.

Rapholo

S. F.

Makhubele

J. C.

(2019). Forensic interviewing techniques in child sexual abuse allegations: Implications for the South African context. Global Journal of Health Science, 11(6), 53–63. https://doi.org/10.5539/gjhs.v11n6p53.

60.

Shoham

(2014). Dat, hiloniut umasoret bemachshava haziborit beIsrael [Religion, secularism and tradition in public thought in Israel]. Studies in the Restoration of Israel, 24, 29–58.

61.

Tishelman

A. C.

Fontes

L. A.

(2017). Religion in child sexual abuse forensic interviews. Child abuse & neglect, 63, 120–130.

62.

Tulving

(1983). Elements of episodic memory. Oxford University Press.

63.

Sharabi

Kay

(2021). The relative centrality of life domains among secular, traditionalist and Ultra-Orthodox (Haredi) men in Israel. Community, Work & Family, 24(1), 60–76. https://doi.org/10.1080/13668803.2019.1675591.