International researchers and child protection service workers beliefs about child sexual abuse disclosure and statement validity

ABSTRACT

How child victims of sexual abuse disclose their experiences is contested among experts. We surveyed international researchers (N = 199) and child protection service workers (N = 267) on their beliefs regarding how victims of child sexual abuse cope with and disclose their experiences, and how these disclosure patterns affect the validity of statements given by maltreated children. We found some points of disagreement among experts related to, for example, recantation and denial frequency. However, there were also certain points of agreement such as that children often delay disclosure, and that false denials can affect statement validity (i.e. the accuracy and truthfulness of a statement) negatively. However, the majority of both expert groups agreed that statements made after denial or recantations should be admissible as evidence in court. Expert witnesses could take areas of agreement into account when giving testimony in child sexual abuse cases.

KEYWORDS:

• Child sexual abuse
• disclosure
• expert beliefs
• statement validity
• memory

Child sexual abuse (i.e. CSA) is a serious problem occurring on a global level, with international prevalence rates of 8-31% for girls and 3-17% for boys under the age of 18 (Barth et al., 2013; see also Finkelhor et al., 2014; Moody et al., 2018; Stoltenborgh et al., 2015). When such cases are eventually referred to court, psychologists and other experts sometimes act as expert witnesses. Their job is then to inform the judge or jury on how children who truly have been abused might come forward about such cases, and how valid the statements of an alleged victim might be. These judgements can be based on how alleged victims were interviewed, or how they came forward about their experiences (Kovera & Borgida, 1997).

Summit (1983) was one of the first psychiatrists who described how CSA victims might come forward and eventually report on their negative experiences. By accumulating his clinical experience with victims of abuse and examining current literature, Summit (1983) described that sexually abused children would often be met with disbelief by non-abusing family members if they attempted to come forward. Moreover, their traumatic experiences would lead to certain psychological and emotional turmoil such as self-blame and shame. To accommodate this turmoil, Summit (1983) described that sexually abused children tend to engage in certain behaviours of secrecy: They might delay disclosing their experiences, (falsely) deny them when asked by others, and even retract allegations they made at an earlier stage.

Since Summit (1983) published his views on how CSA victims disclose abuse, controversy on the scientific foundation of these behavioural and emotional response patterns has been sparked (see for instance London et al., 2020). To this end, researchers have examined the scientific basis for these different possible disclosure patterns. While these attempts seem to have led to agreement about some disclosure responses (e.g. delayed disclosure), in the literature, researchers continue to disagree with each other about some of the other behavioural patterns (London et al., 2020; Lyon et al., 2020; Summit, 1983). Controversial disclosure patterns include the occurrence of false denials and recantations of abuse.

A pernicious situation occurs when researchers seem to disagree on the disclosure process of CSA victims and when such disagreements play out in legal cases (e.g. London et al., 2020; Lyon et al., 2020; New Jersey vs J.L.G., 2018). For example, if the hired expert believes denials do not occur often among real abuse victims, the expert might be more likely to deem an alleged victim to be denying because they really were not abused. However, a hired expert faced with the same alleged victim who believes that it is not uncommon for truly abused children to (falsely) deny abuse might be more likely to continue interviews. This is in line with research showing that interviewers who have strong reasons to believe the child has been abused might turn to more suggestive questions when faced with uncooperative children (Hershkowitz et al., 2006). At the moment, we do not know the extent to which researchers disagree and agree about the specific prevalence of certain disclosure patterns. It is an empirical question what opinions they hold regarding how such patterns might affect validity of testimonies and where possible agreement exists.

Agreement is important not just to evaluate the current state of a field, but also because agreement in the relevant, specific scientific field – which must be defined – is one of the guidelines trial judges should use in countries such as the USA to evaluate whether expert witness evidence should be admissible (Daubert v. Merrell Dow Pharmaceuticals, Inc., 1993). The other Daubert guidelines establish that the underlying theory or technique (1) should be falsifiable, (2) has been subjected to peer review, and (3) has a known error rate (Daubert v. Merrell Dow Pharmaceuticals, Inc., 1993). Until 1993, agreement in the scientific field was the only guideline for admissibility of expert evidence in the US (Frye v. United States, 1923). Another group of experts important to the debate of how children disclose abuse experiences are professionals who interview children of alleged sexual abuse. In many countries (e.g. the Netherlands, the US, Canada), professionals employed by child protective services (CPS) tend to conduct the first interviews with children in cases in which abuse is suspected (Bala et al., 2001; Child Welfare Information Getaway, 2020; Children’s Bureau, 2020; Cronch et al., 2006; Erens et al., 2020). These CPS employees are also responsible for deciding whether follow-up actions – like reporting the case to the police – are necessary (Erens et al., 2020; Levenson & Morin, 2006). Nevertheless, their beliefs about how truly abused children will disclose in interviews, and how disclosure patterns reflect and affect the validity of statements might then guide their decisions on whether a case should be forwarded to the police or not. Indeed, there is much research showing that beliefs can guide behaviours (e.g. Ajzen & Fishbein, 1972; Armitage & Christian, 2003; McEachan et al., 2011). Of note, in case there are suspicions of criminal child abuse (e.g. CSA), these are always reported to the police (Erens et al., 2020).

Moreover, it should be acknowledged that the specific role of experts can differ between countries. In some countries, such as the Netherlands (Erens et al., 2020), several states in the US (Child Welfare Information Getaway, 2020; Children’s Bureau, 2020) and Canada (Bala et al., 2001), child protection service (i.e. CPS) workers directly conduct interviews with alleged child victims. However, in other countries, such as Australia, it might primarily the job of the police to conduct such interviews (e.g. the Child Protection and Investigation Units in Australia, Queensland Police, 2021).

Thus, the aim of the current study is to investigate the beliefs of both researchers in the field of CSA and memory, and CPS employees regarding the different disclosure patterns of children who were sexually abused and their possible effects on memory reports and statement validity (i.e. the accuracy and truthfulness of statements given by the victim).

CSA disclosure patterns

Delayed disclosure

Delayed disclosure is a term used to refer to the situation when someone who has experienced a negative event does not come forward about it right away, for example, to a family member, a friend, or to officials, but only after some time has passed (London et al., 2005). Delayed disclosure implies that the person eventually does come forward after this period of silence. Evidence for this idea comes from adult retrospective surveys in which adults are asked whether they ever experienced sexual abuse as a child, and if so, whether they ever told anyone about this abuse (Alaggia et al., 2019; Priebe & Svedin, 2008). In 2005 and 2008, London and colleagues reviewed the results of such surveys. They found that 55-69% of adults who self-reported childhood sexual abusive experiences never told anyone about their experiences until adulthood (i.e. they delayed disclosing the abuse for many years). Similarly, a recent longitudinal study on childhood sexual abuse victims with a 20 year follow-up found that deferred disclosure (i.e. delayed disclosure following an initial period of denial) was not uncommon (17%, 17/99; Hartman et al., 2023).

Delaying disclosure of abuse experiences preserves the secrecy around victims’ adverse experiences and thus possibly exposes the victim to repeated abuse (Romeo et al., 2018). Overall, there is scientific support and general agreement among researchers that delayed disclosure is common among CSA victims (see also: London et al., 2020; Lyon et al., 2020).

False denial

Debate exists concerning the frequency with which false denials occur. A false denial transpires when someone, for example, a victim of abuse, denies that a certain action or event occurred (to them, e.g. ‘No one touched me’.), even though in reality it did (Bücken et al., 2022a; Lyon, 2007). There are myriad reasons why a victim might falsely deny abuse (Romeo et al., 2018). Denials are sometimes employed as a coping mechanism or because of external threats and pressures (Romeo et al., 2018). A reason why opinions on how often victims use such false denials differ might be the difficulty of obtaining scientific evidence for these prevalence rates. Researchers in the field have stated that it is difficult to study denials and recantations in victims of abuse in a valid way because the ground truth is often not known, and there are issues of suspicion and substantiation bias (Lyon et al., 2020). Suspicion bias refers to the bias that allegations oftentimes arrive, and children come to the attention of authorities, because they have already disclosed that something happened to them. Substantiation bias occurs when substantiation is dependent on a child’s disclosure, for example when there is a lack of other evidence. Both of these biases inflate disclosure rates and decrease denial rates (Lyon et al., 2020). The ground truth problem, for example, renders it difficult to estimate whether a denial is a truthful denial (i.e. no abuse took place) or a false denial (London et al., 2005; Lyon, 2007).

Researchers disagree whether false denials of CSA occur frequently or rather rarely. On the one hand, Lyon (2007) attempted to forego substantiation and suspicion bias by examining a population of children where external medical evidence was available, but children had not disclosed that anything happened to them. Specifically, he examined studies in which children were diagnosed with sexually transmitted diseases but had not disclosed any abuse previously and found an average false denial rate of 58%. On the other hand, London and colleagues (2005; 2020) examined studies of all children who were interviewed by authorities and found an average of 36% (4%−76% range) denials, 15% when the authors only included cases they deemed to be verified abuse cases.¹

Recantation

Recantations occur when someone who has previously disclosed that they have experienced abuse takes this claim back and then denies (i.e. recants) experiencing the events (Malloy et al., 2007). There are different reasons why a victim of childhood sexual abuse might recant earlier claims. These include not feeling supported, fears of shifts in the family dynamic, external pressures, and feelings of guilt (Hartman et al., 2023; Malloy et al., 2007). The prevalence of such recantations is also disputed among the scientific community (London et al., 2020; Lyon, 2002). Some scholars view recantations of authentic abuse as occurring quite rarely (e.g. London et al., 2005 found an average rate of 13%, but range of 4-27% in their review of eight studies). Other scholars (e.g. Lyon et al., 2020; Malloy et al., 2007, p. 2016) have argued that recantations are more common in certain subpopulations of sexually abused children than others. Indeed, in their filial dependency theory, Malloy and colleagues (2007, 2016) explained that, in their sample, sexually abused children were more likely to recant if they were not believed or supported by non-abusing family members as pitted against children who were believed by non-abusing family members.

Beliefs about disclosure

Several surveys have evaluated beliefs about disclosure of CSA in various populations (e.g. expert and non-expert). Table 1 summarises findings regarding endorsement of non- and delayed disclosure, false denials, and recantations, as well as consequences for statement validity. Specifically, data from four surveys are combined: a survey by McGuire and London (2017) with an undergraduate student sample, a survey by Quas et al. (2005) with a sample of 148 jurors and 169 juror eligible adults, work by Kovera and Borgida (1997) examining opinions of 156 members of the International Society for Traumatic Stress Studies, and Pelisoli et al. (2015) examining the opinions of CSA professionals in the US (i.e. social and case workers, physicians, psychologists, and nurses) and Brazil (i.e. psychologists) and undergraduate college students (i.e. non-professionals).

Table 1. Previous survey research on beliefs surrounding child sexual abuse cases.

Survey	Pelisoli et al. (2015)	Kovera and Borgida (1997)	McGuire and London (2017)	Quas et al. (2005)
Sample	30 US psychologists, 158 other CSA professionals*, 457 undergraduate students	156 experts in the field of CSA	670 undergraduate students	148 jurors & 169 jury eligible undergraduates
Disclosure Patterns
non-disclosure	90% (27/30) US psychologists, 94% undergraduate students (430/ 457), 94% other professionals (149/ 158)	/	43% (287/ 669)	/
delayed disclosure – children do not report the abuse to someone else right away	87% (26/30) US psychologists, 89% (141/158) other professionals, 70% (320/ 457) undergraduate students	62.% (97/156)	32% (213/669)	84% (266/317)
Denial	/	/	86% (576/669)	43% (136/317) if asked by a trusted adult
Recantation	80% (24/30) US psychologists, 79% (125/158) other professionals, 74% (338/457) undergraduate students said 25% of victims recant	/	65% (432/669) said CSA victims tend to recant	/
Emotional/ Psychological Consequences
loving abusive parent	/	60% (94/157)	/	/
emotional component: self-doubt, self-blame, shame	/	/	/	/
Consequences for validity/ memory
Of inconsistencies, generally	/	/	/	29% (92/317) thought inconsistencies indicate the report is false
disclose more details over time as a sign accounts are true	/	/	/	52% (165/ 317)
recantation as a sign that initial statements are false	/	<1% (1/156)	/	/

Note. ‘Other non-professionals’ include: US case and social workers, nurses, and physicians, and Brazilian psychologist. Percentages indicate participants’ agreement with statements.

Table 1 shows that opinions of child abuse professionals (e.g. psychologists) are relatively in agreement that non- and delayed disclosure is common among CSA victims. However, depending on the specific professional sample, year of data collection, and compared with non-professional undergraduate samples, opinions regarding the occurrence of different disclosure patterns differed. Specifically, non-professionals seemed to think that recantations are common, but delayed disclosure somewhat less so (see Table 1).

Moreover, Table 1 shows that while several surveys have already investigated the issue of beliefs surrounding CSA disclosure among some expert (and non-expert) groups, these have only covered certain elements. For example, beliefs about how disclosure patterns are related to statement validity in court have been covered very scarcely in these surveys. This issue of statement validity is important because of the value attached to alleged victims’ testimonies in court. Research on the effects of different disclosure patterns on memory and the validity of statements is limited, but research that is available shows that traumatic events are usually well remembered (Goldfarb et al., 2019). Of course, regular forgetting mechanisms occur (e.g. due to lack of rehearsal and attempting not to think about the event) when disclosure is delayed (Dodier & Tomas, 2019) and it is known that memories fade over time (e.g. Goodman et al., 2019; Greenhoot et al., 2005). Nevertheless, when the memory is eventually triggered and disclosed spontaneously without suggestive influences, even traumatic memories disclosed after a longer delay might be remembered accurately (Goldfarb et al., 2019; Goodman et al., 2019).

Next to regular forgetting that might set in if disclosure is delayed for a longer time period, disclosure patterns including a lie (i.e. false denials) might negatively affect memory. In fact, research to date has repeatedly shown that when someone first falsely denies that they witnessed something, and then comes forward with the truth, this initial denial can have detrimental consequences on memory (for a review: Otgaar & Baker, 2018; see also: Battista et al., 2021; Otgaar et al., 2014; Romeo et al., 2019).

Another limitation of previous surveys relates to the target participants. Even though CPS employees oftentimes are those who complete the first formal interviews (in the Netherlands and the United States, for example; Child Welfare Information Getaway, 2020; Children’s Bureau, 2020; Erens et al., 2020), no research has investigated how they view disclosure of CSA and its consequences on statement validity. However, in the Netherlands, for example, depending on the specific type of case, CPS workers are oftentimes those who decide whether a case should be investigated further and passed on to the police (Erens et al., 2020). CPS workers in the United States of America are also very closely involved in the processing of alleged child abuse cases. Indeed, although exact procedures vary by state in the United States of America, when concerns regarding possible child abuse or neglect are reported, CPS workers will generally be the first to receive these reports (Child Welfare Information Getaway, 2020; Children’s Bureau, 2020). Then, the CPS worker screens the report on whether an official investigation is needed or not, and may also refer the person issuing the report to law enforcement for additional help. When the CPS worker screens a report ‘in’ (i.e. an investigation is started), the caseworker themselves may interview the child, the parents, and other relevant contacts (e.g. doctors) in order to make a decision on whether the allegations are substantiated or not (Children’s Bureau, 2020; Child Welfare Information Getaway, 2020). The CPS may then initiate court action, if deemed necessary. Even though CPS workers are oftentimes very directly involved in alleged abuse cases and in interviewing children, there is very limited research involving this group. There has been some work on the knowledge of CPS ‘Safe Home’ agency employees in the Netherlands regarding how memory works (Erens et al., 2020). This work has revealed that CPS professionals hold some beliefs in line with memory research, but also endorse memory concepts such as unconscious repression of traumatic memory, which is largely scientifically unfounded (Loftus & Ketcham, 1996; McNally, 2005; Otgaar et al., 2022; Patihis et al., 2014). However, no elements related to CSA disclosure and statement validity were assessed in this research.

The current study

The current study examined disclosure patterns of CSA victims by surveying (a) CPS employees (who often conduct the first interviews with such children) and (b) CSA (and memory) researchers who might evaluate statements and their validity if a case is referred to court. In this survey, we investigated the endorsement of (1) different disclosure patterns of CSA and (2) beliefs about how these affect statement validity. Because research into the cognitive effects of some of these different disclosure patterns is limited (i.e. especially regarding recantation), the latter has been included as an exploratory aim. Nevertheless, in line with recent work on lying and memory by Riesthuis and colleagues (2021) we expected that participants correctly believe that false denials can have negative consequences on memory (i.e. forgetting of details). Moreover, as an exploratory aim, beliefs about the emotional reactions to CSA were examined [e.g. ‘Children who have suffered intra-familial sexual abuse (i.e. by a family member) often develop feelings of self-blame or self-doubt’] and results can be found in the supplementary material. The study was preregistered on the Open Science Framework (OSF; https://osf.io/8jwtb/?view_only = 0063e7fb41dd43e794a6e58451e04d9b).

Materials & methods

Participants

Participants were international professionals employed at CPS agencies (i.e. we had originally planned to recruit these at Safe Home in the Netherlands but additionally were able to reach a more international sample through social media) and research experts in areas related to CSA and statement validity. We performed sample size calculations using the Qualtrics survey sample size calculator (https://www.qualtrics.com/blog/calculating-sample-size/), based on our original plan to survey CPS workers at Safe Home in the Netherlands. With a confidence level of 95%, estimated population size of the Safe Home employees of 2080, and a 10% margin of error, our ideal sample size was N = 92. We expected to receive responses from a similar amount of researchers in total for the relevant areas of CSA, developmental psychology, statement validity, and memory. Participants had the possibility to receive a 5€ gift voucher. We received ethical approval for the study from the KU Leuven social and societal ethics committee [SMEC; code: G-2021-3833-R2]. All participants received an information letter for the study as the first page of the questionnaire and provided informed consent before being able to access the questionnaire.

In total, we collected data from 573 participants. Of these, 229 were international researchers, 295 were CPS workers, and 21 described their profession as ‘other’. Because some participants who picked ‘other’ as their profession worked in areas unrelated to our areas of interest, data from these people were excluded from our analyses. Moreover, we excluded participants who did not finish the survey (n = 29), and those who failed both of our attention checks (n = 29). Thus, the final sample constituted of 199 researchers (n_Male = 100, n_Female = 98, n_{PreferNotToSay}= 1) and 267 CPS workers (n_Male = 109, n_Female = 158). Researchers were on average 34.6 years old (SD = 8; range = 22 - 75 years old), and 66% of them (131/199) had previously worked as an expert witness on a CSA case.² Of these expert witnesses, 46% (60/131) had worked on more than 20 court cases involving alleged CSA. Table 2 provides an overview regarding the background of researchers involved in our sample. Our CPS sample was on average 31.98 years old (SD = 7.56; range = 18–60 years old) and was composed of social workers (32%, 85/267), psychologists (21%, 57/267), nurses (24%, 63/267), doctors (16%, 42/267), and ‘others’ (7%, 20/267), for instance behavioural scientists. The majority of these had between 1 and 6 years of experience in their field (69%, 185/267), but some also had more years of experience (7-10 years: 18%, 47/267; 11–15 years: 8%, 21/267; more than 15 years: 5%, 14/267). Moreover, of our CPS worker sample, 85% (226/267) indicated that they regularly conduct interviews with children for whom abuse is suspected. CPS workers who indicated that they do regularly conduct interviews with children in alleged abuse cases were asked how many of these cases they have approximately worked on. Most participants in this sample had worked on several CPS cases (see Table 2). Table 2 further provides an overview of the countries of employment of our participants, divided into researchers and CPS workers.

Table 2. Participants’ Background

	Researchers		CPS Workers
	n	%	n	%
Years of Experience
1–3 Years	57	29	107	40
4–6 Years	46	2	78	29
7 - 10 Years	29	15	47	18
11–15 Years	17	9	21	8
>15 Years	50	25	14	5
Total	199		267
Number of Expert Witness Cases (Researchers)/alleged abuse cases (CPS workers) worked on
1	3	2	20	9
2–5	40	31	87	39
6–10	18	14	46	20
10–15	4	3	23	10
15–20	6	5	14	6
>20	60	46	36	9
Total	131		226
Country of Employment
The Netherlands	7	4	43	16
Belgium	0	0.0	4	2
UK	18	9	20	8
USA	114	57	175	66
Australia	35	18	15	6
Germany	2	1	2	1
Italy	1	1	1	0
France	1	1	4	2
Spain	2	1	0	0
Sweden	2	1	0	0
Canada	6	3	3	1
Other	11	6	0	0
Total	199		267

Procedure and materials

A link with a Qualtrics survey was distributed via email to employees of Safe Home in the Netherlands, via email to members of the European Association for Psychology and Law (EAPL) and the Society for Applied Research in Memory and Cognition (SARMAC), and via social media (i.e. Twitter/X). Participants received a questionnaire with items asking about the frequency of each of the disclosure patterns described above, the effect of each disclosure pattern on beliefs surrounding validity of the statements, and whether they should be admissible in court (see supplementary material for the full questionnaire). Specifically, the questionnaire was divided into five parts: (1) exploratory questions about the emotional component related to CSA disclosure, (2) delayed disclosure, (3) false denials, (4) recantations, and (5) general questions related to disclosure and statement validity. Questions were based on the literature reviewed above (e.g. Goldfarb et al., 2019; London et al., 2005, 2020; Lyon, 2007; Lyon et al., 2020; Malloy et al., 2007, 2016; Otgaar et al., 2014, 2020; Riesthuis et al., 2022; Summit, 1983, 1993). Furthermore, some items used in or based on previous questionnaires about CSA were included (i.e. from Kovera & Borgida, 1997; McGuire & London, 2017; Pelisoli et al., 2015; Quas et al., 2005; Riesthuis et al., 2022), as marked specifically in the supplementary material. Items were predominantly endorsement questions (e.g. ‘When children disclose sexual abuse by a family member to a non-abusing parent, they are often faced with disbelief’, answer options: ‘Agree (5); somewhat agree (4); neither agree nor disagree (3); somewhat disagree (2); disagree (1)’) and frequency questions (e.g. ‘Sexually abused children (falsely) deny that they were abused when asked by a family member or friend’; answer options: ‘Never (1); rarely (2); sometimes (3); often (4); always (5)’). All materials, as well as supplementary materials, analyses and the full anonymized dataset, can be found on OSF (https://osf.io/8jwtb/?view_only = 2cb15ce851464e8eb88a5b1639120ac7). Data collection took place from October 2021 until February 2022.

We predominantly used frequency analyses and descriptives to analyse our data. Additionally, we conducted exploratory analyses comparing results from the two expert groups (i.e. CPS workers and researchers) using chi-square and independent samples t-tests, depending on the variables. We used JASP for all statistical analyses.

Results

Our results are presented in three parts: (1) behavioural components divided into (a) delayed disclosure, (b) recantation, and (c) false denials (2) the emotional component of CSA disclosure, and (3) general questions regarding disclosure. For each section, the exact questions and their endorsement for both of our samples are presented in a table in the supplementary material. For the purpose of our data analysis, we have collapsed our agree–disagree scores from endorsement questions into three categories: agree, neither agree nor disagree, and disagree. Hence, we collapsed ‘agree’ and ‘somewhat agree’ into one category, and ‘disagree’ and ‘somewhat disagree’ into another (e.g. see Riesthuis et al., 2022). We report only selected results in text in order to keep the paper focused. Further result tables can be found in the supplementary material. If not indicated otherwise, the comparisons drawn between the frequency scores of CPS works and researchers are drawn based on descriptive statistics. The questionnaire results reported in the paper are also summarized in Tables 3 and 4, and all results can be found in Table format on OSF (https://osf.io/8jwtb/?view_only = 2cb15ce851464e8eb88a5b1639120ac7).

Table 3. Survey Results for Items with Response Options ‘Never’ (1) to ‘Always’ (5).

		Researchers (n = 199)											CPS workers (n = 267)
		Never		Rarely		Sometimes		Often		Always			Never		Rarely		Sometimes		Often		Always
	Statements	n	%	n	%	n	%	n	%	n	%	M (SD)	n	%	n	%	n	%	n	%	n	%	M (SD)
False Denials	Sexually abused children (falsely) deny that they were abused when asked by a family member or friend.	5	2.5	26	13	82	41	1	36	15	8	3.3 (0.9)	2	1	71	27	105	39	73	27	16	6	3.1 (0.9)
	Sexually abused children (falsely) deny that they were abused when interviewed by authorities (e.g. child protective services, the police, medical staff).*	0	0.0	43	22	78	39	50	25	28	14	3.3 (1.0)	6	2	66	25	120	45	61	23	14	5	3.0 (0.9)
	Children who denied might eventually come forward with claims of abuse if interviewed according to research based practices.	1	0.5	40	20	91	46	49	26	18	9	3.2 (0.9)	11	4	81	30	125	47	34	13	16	6	2.9 (0.9)
	Children who denied might eventually come forward with claims of abuse if interviewed suggestively.	2	1	39	20	93	47	48	24	17	9	3.2 (0.9)	5	2	70	26	140	52	37	14	15	6	3.0 (0.8)
Recantations	Sexually abused children who tell someone about their abuse will change their story later and say that they were not sexually abused (i.e. recant their claims) when asked by family members or friends.	1	0.5	32	16	100	50	50	25	16	8	3.2 (0.8)	8	3	75	28	125	47	50	19	9	3	2.9 (0.9)
	Sexually abused children who tell someone about their abuse will change their story later and say that they were not sexually abused (i.e. recant their claims) if asked by a formal authority.*^2	3	1.5	41	21	92	46	38	19	25	13	3.2 (1.0)	9	3	76	29	118	44	53	20	11	4	2.9 (0.9)

Note. * t(403.86) = 3.16, p = .002, d = 0.3; *² t(464) = 3.228, p = .001, d = 0.3

Table 4. Survey results for disagree (1) – agree (5) items.

		Researchers (n = 199)							CPS workers (n = 267)
		Disagree		Neither Agree nor Disagree		Agree			Disagree		Neither Agree nor Disagree		Agree
	Statements	n	%	n	%	n	%	M (SD)	n	%	n	%	n	%	M (SD)
Delayed Disclosure	When sexually abused children do not tell anyone right away but after a delay, they might forget and recollect less true details of their experience.	34	17	29	15	136	68	3.8 (1.2)	63	24	45	17	159	60	3.5 (1.2)
	When sexually abused children come forward with the abuse in adulthood after not telling anyone for a long time, the validity of their statements is negatively affected.	35	18	27	14	137	69	3.8 (1.2)	44	17	55	21	168	63	3.6 (1.0)
	When there is a long delay between when the alleged abuse occurred in childhood and when a person comes forward about this abuse, then these statements should be admissible in a court setting.	17	9	19	10	163	82	4.3 (1.0)	33	12	53	20	181	68	3.9 (1.0)
False Denial	When sexually abused children deny this abuse, they might forget details about their experience.	59	30	36	18	104	52	3.3 (1.2)	88	33	68	26	111	42	3.1 (1.1)
	When sexually abused children deny this abuse, they might forget details about previous interviews in which they denied the abuse.	37	19	34	17	128	64	3.6 (1.1)	67	25	67	25	133	50	3.3 (1.1)
	When sexually abused children deny this abuse, they might remember and report more false details about their experience compared to victims who did not deny their experience.	54	27	46	23	99	50	3.3 (1.2)	69	26	63	24	135	51	3.3 (1.1)
	When children do eventually come forward with statements about abuse after first denying the abuse, the validity of their statements is negatively affected.*	31 (/194)	16	29 (/194)	15	134 (/194)	69	3.8 (1.1)	33 (/226)	15	46 (/226)	20	147 (/226)	65	3.7 (1.0)
	When children do eventually come forward with statements about abuse after first denying the abuse, their statements should be admissible as evidence in court.*	15 (/194)	7	16 (/194)	8	163 (/194)	84	4.4 (1.0)	20 (/226)	9	27 (/226)	12	179 (/226)	67	4.0 (0.9)
Recantations	In some contexts (e.g. if the abuser is a parent or if the non-abusing parent is unsupportive) more children who were sexually abused will recant their allegations compared to other contexts (e.g. supportive non-abusing parent, the abuser is a stranger).	35	18	43	22	121	61	3.6 (1.1)	44	16	71	27	152	57	3.5 (1.0)
	When children disclose authentic abuse and then recant their allegations, they might forget details of their experience.	57	29	27	14	115	58	3.4 (1.3)	109	41	73	27	85	32	2.9 (1.1)
	Children who retract their stories about sexual abuse were probably lying in the first place.	63	32	105	53	31	16	2.7 (1.3)	106	40	104	39	57	21	2.7 (1.2)
	When children disclose abuse and then recant their allegations, the validity of their statements is negatively affected.	31	16	24	12	144	72	3.9 (1.1)	56	21	49	18	162	61	3.6 (1.1)
	When children recant previous allegations, their statements should be admissible in court.	12	6	20	11	167	84	4.2 (0.9)	28	11	72	27	167	63	3.8 (1.0)
Emotional Component	When children disclose sexual abuse by a family member to a non-abusing parent, they are often faced with disbelief.	4	2	14	7	181	91.0	4.5 (0.7)	9	3	12	5	246	92	4.5 (0.8)
	Children who have suffered intra-familial sexual abuse (i.e. by a family member) often develop feelings of self-blame or self-doubt.	6	3	7	4	186	93.5	4.6 (0.8)	17	6	15	6	235	88	4.3 (0.9)
	Children who have suffered intra-familial sexual abuse often develop feelings of shame.	3	2	8	4	188	94.5	4.7 (0.6)	5	2	16	6	246	92	4.5 (0.7)
	Children who have suffered intra-familial sexual abuse try to keep the abuse secret from others.	5	3	13	7	181	91.0	4.5 (0.8)	14	5	12	5	241	90	4.3 (0.8)
	Children who have suffered intra-familial sexual abuse make efforts to accept the abuse.	97	49	33	17	69	34.7	2.7 (1.4)	91	34	60	23	116	44	3.1 (1.2)
	If a child is sexually abused by a parent, the child will still show love for that parent.	78	39	32	16	89	44.7	3.0 (1.4)	82	31	51	19	134	59	3.2 (1.3)
General Statements	Inconsistencies in a child’s report of sexual abuse (e.g. by initially denying and then coming forward or by disclosing and then recanting) indicate that the report is false.	106	53	24	12	69	34.7	2.7 (1.4)	102	38	74	28	91	34	2.9 (1.2)
	When a child’s description of sexual abuse is disclosed over time, with more details being reported each time the child is interviewed, this clearly indicates that the child’s description is true.	25	13	39	20	135	67.8	3.9 (1.2)	27	10	85	32	155	58	3.8 (1.0)

Note. Means and standard deviations are based on the original 5 point scale for the endorsement questions. *: these questions were only shown to participants when they indicated that they thought false denials occurred ‘rarely’, ‘sometimes, ‘often’, or ‘always’ when sexually abused children are asked about their experiences by (a) authorities or (b) family members/ friends. Therefore, the size of the sample is different there, as indicated in the table in parentheses.

Behavioural components

Delayed disclosure

Many researchers [65% (130/199)) and CPS workers (68% (181/267)] believed that most children who are sexually abused would tell someone at some point in their lives, whereas the other participants thought that such victims would never come forward. Only a small number of people believed that sexually abused children who do come forward would tell right away [researchers: 11% (21/199); CPS workers: 17% (46/267)], whereas most believed that there would be a delay to reporting [researchers: 89% (178/199); CPS workers: 83% (221/267)]. Nevertheless, there was some disparity among both researchers and CPS workers regarding when such victims would come forward [CPS workers: within 1 year: 23% (61/267); before adulthood: 28%, (76/267); in adulthood: 31% (84/267); researchers: within 1 year: 22% (44/199); before adulthood: 30% (59/199); in adulthood: 38% (75/199)].

Delayed Disclosure and Statement Validity. We also asked participants about their opinions regarding how such delayed disclosure might affect memory and statement validity. To this end, the majority of researchers and CPS workers agreed that a child victim might have forgotten some true details of their experience when they disclose after a delay [researchers: 68% (136/199); CPS workers: 60% (159/ 267)], and that the validity of their statements might be negatively affected by such a delay [if the child comes forward in adulthood: researchers: 69% (137/ 199); CPS workers: 63% (168/267)] (see Table 3). The majority of both groups of participants agreed that when there is a long delay between the alleged abuse and a person coming forward, the statements should be admissible in a court setting [researchers: 82% (163/ 199), CPS workers: 68% (181/267)]. Of note, this majority was (descriptively) larger for researchers than CPS workers.

False denial

We asked participants about how frequently they thought false denials occurred among sexually abused children in both informal (e.g. when asked by a family member or friend) and formal settings (e.g. when asked by a formal authority such as child protective services, the police, or medical staff) Both researchers and CPS workers believed that false denials do happen in both settings – only 3% (5/199) of researchers and 1% (2/267) of CPS workers thought that informal false denials never happen (see Table 4). Moreover, no (0%, 0/199) researchers and only 2% (6/267) of CPS workers thought false denials never happen in formal interview settings. While both groups of participants thus agreed that false denials occur among abused children, opinions with regards to the specific frequency of false denials in formal settings were scattered among both researchers [rarely: 22% (43/199); sometimes: 39% (78/199); often: 25% (50/199)] and CPS workers [rarely: 24% (66/267); sometimes: 45% (120/267); often: 23% (61/267)]. Of note, 14% (28/199) of researchers and 5% (14/267) of CPS workers even stated that false denials would always occur in this setting. We were also interested in whether researchers and CPS workers differed statistically significantly in how often they believed sexually abused children would falsely deny in formal settings. To this end, we treated the response point as a Likert scale ranging from never (1) to always (5) and ran a Welch t-test to explore whether researchers and CPS workers differed in how often they believed false denials would occur in formal interviews. Researchers on average believed that false denials occurred more frequently (M = 3.32, SD = 0.97) as compared with CPS workers (M = 3.04, SD = 0.88), t(403.86) = 3.158, p = .002, d = 0.3.³

Generally, a large proportion of participants [researchers: 46% (91/199); CPS workers: 47% (125/267)] believed that children who initially denied might sometimes eventually come forward with claims of abuse if interviewed according to research-based practices (i.e. according to current interviewing guidelines which are based on research concerning effective and ethical interviewing). Some participants also thought that this would happen rarely [researchers: 20% (40/199); CPS workers: 30%, (81/267)] or often [researchers: 26% (49/199); CPS workers:13% (34/267)], but only few endorsed the extremes never [researchers: 1% (1/199); CPS workers: 4.% (11/267)] and always [researchers: 9% (18/199); CPS workers: 6% (16/267)]. Results were very similar when we asked participants how frequently they thought that children who initially denied might eventually come forward with claims of abuse if interviewed suggestively (i.e. not according to evidence-based guidelines; see for example Blasbalg et al., 2021). Specifically, about half of the researchers (47%, 93/199) and CPS workers (52%, 140/267) believed that children might sometimes come forward with abuse claims after they had initially denied abuse, if interviewed suggestively (see also Table 3).

False Denial and Statement Validity. We asked participants how they thought false denials might affect memory and the validity of statements eventually given by CSA victims. Overall, participants’ beliefs were somewhat scattered regarding statements of how false denials would affect statement validity (i.e. details forgotten, details falsely remembered, and details regarding previous interviews), and around a fourth of CPS participants indicated that they neither agreed nor disagreed that false denials would affect memory reports [25% (68/267)] (see Table 3). However, around half of the participants agreed that victims might forget details of (1) their experience [researchers: 52% (104/199); CPS workers: 42% (111/267)] and (2) previous interviews in which they denied the abuse [researchers: 64% (128/199); CPS workers: 50% (133/267)] and that victims who denied previously might report more false details of the experience compared to victims who did not deny [researchers: 50% (99/199); CPS workers: 50% (135/267)]. Generally, the majority of researchers (70%, 134/194)⁴ and CPS workers (65%, 147/226) stated that statement validity would be negatively affected after false denials. Finally, the majority of both groups indicated that statements made after initial denials should be admissible as evidence in court [researchers: 84% (163/194); CPS workers: 67% (179/226)]. Of note, this majority was (descriptively) bigger for researchers than CPS workers. Moreover, interestingly, 7% (15/194) of researchers and 9% (27/226) of CPS workers disagreed that statements made after denials should be admissible as evidence.⁴

Recantations

Last, we asked participants how often they thought victims of child sexual abuse would recant claims they initially made. Similar to false denials, only few participants stated that recantations would never happen in an informal (researchers: 1%, 1/199; CPS workers: 3%, 8/267) or formal setting (researchers: 2%, 3/199; CPS workers: 3%, 9/267) (see Table 4). Indeed, there was a general agreement that recantations do occur. However, opinions with regards to how often recantations would happen in formal settings were heterogenous among both researchers (rarely: 21%, 41/199; sometimes: 46%, 92/199; often: 19%, 38/199; always: 13%, 25/199) and CPS workers (rarely: 29%, 76/267; sometimes: 44%, 118/267; often: 20%, 53/267; always: 4%, 11/267). So, it seems that a higher percentage of researchers, as compared with CPS workers, indicated that abused children would always recant earlier statements. We treated the response point as a Likert scale ranging from never (1) to always (5) and ran a Students’ t-test to statistically compare if researchers and CPS workers statistically differed in how often they believe recantations occur in formal settings. This difference was statistically significant, t(464) = 3.228, p = .001, d = 0.3, with researchers indicating on average a higher believed prevalence of recantations (M = 3.21, SD = 0.96) than CPS workers (M = 2.93, SD = .88).

We further explored whether responders’ scores on their belief regarding the frequency with which denials take place and the frequency with which recantations take place would correlate. We indeed found a statistically significant positive correlation of scores on the two items: r (464) = .492, p < .001. Additionally, we found that the majority of participants [researchers: 61% (121/199); CPS workers: 57% (152/267)] agreed that recantations are more likely in some contexts (e.g. the abuser is a parent) than others (e.g. the abuser is a stranger, see Table 4). Interestingly, a minority of researchers (18%, 35/199) and also CPS workers: (17%, 44/267) disagreed with this statement, thus indicating a belief that recantations occur independently of the specific circumstances.

Recantations and Statement Validity. By and large, while the majority of researchers (58%, 115/199) agreed that victims of abuse might forget details of their experience after recantations, descriptively more CPS workers disagreed (41%, 109/267) than agreed (32%, 85/267) with this statement (Table 4). Moreover, opinions of both researchers (disagree: 32%, 63/199, neither agree nor disagree: 53%, 105/199, agree: 16%, 31/199) and CPS workers (disagree: 40%, 106/267, neither agree nor disagree: 39% 104/267, agree: 21%, 57/267) were scattered when we asked whether children who retract stories about sexual abuse were ‘probably lying in the first place’. The majority of both researchers (72%, 144/199) and CPS workers (61%, 162/1267) agreed that when children recant abuse after first disclosing it, the validity of their statements is negatively affected. However, most researchers (84%, 167/199) and CPS workers (63%, 167/267) agreed that even after recantations, children’s statements should be admissible in court.

Emotional component

Both researchers and CPS workers mostly agreed with some emotional components related to CSA disclosure. Specifically, 91% (181/199) of researchers and 92% (246/ 267) of CPS workers agreed with the statement that children are often faced with disbelief when disclosing sexual abuse to a non-abusing parent (Table 4). Moreover, they agreed with that child victims of intra-familiar abuse oftentimes develop feelings of self-blame or self-doubt [researchers: 93% (186/199), CPS workers: 88% (235/267)], feelings of shame [researchers: 94% (188/199), CPS workers: 92% (246/267)], and try to keep the abuse secret from others [researchers: 91% (181/199), CPS workers: 90% (241/267)]. However, both researchers and CPS workers’ beliefs of whether children who are victims of intra-familial abuse (1) make efforts to accept the abuse and (2) still show love for the abusive parent showed less agreement within and between both groups. Indeed, the majority of researchers disagreed [49% (97/199)] with the idea that children who have suffered intra-familial abuse would make efforts to accept the abuse, but a large percentage of them also agreed with this idea [35% (69/199)]. On the other hand, the majority of CPS workers [44% (116/267)] agreed with that children would make efforts to accept the abuse, and 34% (91/267) disagreed. Regarding the item ‘showing love for the abusive parent’, a high percentage of participants disagreed with the statement [researchers 39% (78/199); CPS workers 30.7% (82/267)], but a large percentage also agreed with it [researchers 44.7% (89/199); CPS workers 59.2% (134/267); see Table 3].

General statements

Finally, we presented participants with some general statements regarding inconsistencies and their relation to statement validity. Opinions of both researchers and CPS workers were scattered when we asked whether inconsistencies in a child’s report of sexual abuse (e.g. by initially denying and then coming forward or by disclosing and then recanting) would indicate that the report is false (Table 4). Among the researchers, 53% (106/199) disagreed, 12% (24/199) neither agreed nor disagreed, and 35% (69/199) agreed with the statement. Among CPS workers, 38% (102/267) disagreed, 28% (74/267) neither agreed nor disagreed, and 34% (91/267) agreed with the statement. Additionally, the majority of researchers (68%, 135/199) and CPS workers (58%, 155/267) agreed that ‘when a child’s description of sexual abuse is disclosed over time […] this clearly indicates that the child’s description is true’.

Discussion

The aim of the present study was to investigate the beliefs of professionals (i.e. researchers and child protective service employees) concerning disclosure of childhood sexual abuse and possible consequences for statement validity of different disclosure patterns. Researchers sometimes disagree on certain disclosure patterns concerning abuse (for a review: Bücken et al., 2023). It is vital to understand these areas of agreement and disagreement among researchers. One of the reasons is because these researchers sometimes provide expert testimony in court. It is equally relevant to understand the beliefs regarding disclosure of CPS workers who oftentimes conduct interviews with children in alleged abuse cases. Such interviewers can play an important role in deciding whether a case should proceed to a police investigation or can be ‘dismissed’. Interviewers’ beliefs regarding disclosure and the validity of statements they are faced with help guide this important decision.

One of our most important findings was that some areas of agreement do exist among the experts. Specifically, we found that participants mostly agreed that delayed disclosure is common among abused children, and that both false denials and recantations can in principle happen. Indeed, most participants who did believe that victims would disclose their abuse experiences at some point in their lives, believed that this would happen after some sort of delay. Furthermore, these findings are in line with evidence from adult retrospective surveys (Alaggia et al., 2019; Priebe & Svedin, 2008). Such retrospective surveys also show that CSA victims oftentimes delay disclosing their experiences or might never tell anyone about their abuse at all (Alaggia et al., 2019; Priebe & Svedin, 2008).

Moreover, both researchers and CPS workers agreed that false denials and recantations can happen; almost no participants indicated that they believed these behavioural patterns ‘never’ happened. Thus, even though there is a debate in the literature regarding the frequency with which false denials and recantations occur among true child abuse victims (London et al., 2005, p. 2008; 2020; Lyon, 2007; Lyon et al., 2020), there seems to be a general agreement that it is possible that an abused child falsely denies, or recants earlier disclosures. This result might be related to reviews that do not demonstrate that false denials and recantations never happen, even if they found lower frequencies prevalences than researchers looking at different samples (Eisen et al., 2021; London et al., 2005; London et al., 2020; Lyon, 2007; Lyon et al., 2020).

Even though there was some agreement among experts, there was – as expected – also disagreement. For example, even though there was agreement that CSA victims often delay disclosure (or fail to disclose at all), opinions varied regarding how long such a delay might last. Similarly, there were vastly different opinions regarding the specific frequency with which false denials and recantations might occur. A substantial percentage of participants reported that false denials and recantations happened rarely, but another substantial percentage of participants thought they happened sometimes, often, or even always. Indeed, the apparent disagreement among researchers in the field about the prevalence of false denials and recantations in formal interviews with abuse victims (Bücken et al., 2022) was reflected by our data. Overall, we were able to show that the apparent disagreement regarding the frequency of false denials and recantations is not just carried by some researchers who have published articles in relation to this debate, but widely concerns the entire field.

When comparing our two samples, researchers stated that both false denials and recantations happen more often than CPS workers. This is interesting given that a high percentage of our researcher sample self-reported working as expert witnesses, and that a high percentage of our CPS worker sample indicated that they interview children in alleged CSA cases. Thus, our research participants who oftentimes act as expert witnesses believed that children report abusive experiences less consistently compared with our CPS participants who are directly faced with disclosure. Of course, it is possible that the beliefs of researchers and expert witnesses are biased by the types of cases that they are confronted with or involved in, seen as past experience is a known source of bias (Dror, 2020). It is certainly conceivable that expert witnesses – and thus researchers – are asked to testify in especially complex cases (Reifert, 2011) rather than less complex, straightforward cases. Thus, they might often be confronted with cases in which disclosure is not straightforward but inconsistent. Such bias in case involvement would perhaps not apply to CPS workers, who see and are involved in the full breadth of alleged child abuse cases.

A fraction of researchers (but not so much CPS workers) even believed that child abuse victims would ‘always’ falsely deny (14%; 28/199) and recant (13%, 28/199) in formal interviews. This notion then that all CSA victims would falsely deny their abuse experiences or recant at some point in the interviewing process is not in line with the scientific literature (for a review: Bücken et al., 2022a; Eisen et al., 2021; London et al., 2005; Lyon, 2007). When taking a closer look at this group, we found that half (50%, 14/28) of the researchers who indicated that they thought CSA victims would always falsely deny their experiences in official interviews had worked as expert witnesses in CSA cases.

Moreover, a minority of researchers and CPS workers disagreed that statements made by children after initial denials, or recantations, should be admissible as evidence in court. This is interesting not just because the majority of our participants did think that such evidence should be admissible, but also because this disagreement might have implications for court. If an expert who believes that any statements made after initial denials should not be admissible is hired, their expert testimony would differ greatly from that of other experts. This perhaps would lead to a situation where a victim’s truthful statements could be dismissed by the court.

There are several limitations worth acknowledging. For one, even though we recruited research participants from two professional organisations (the Society of Applied Research in Memory and Cognition and the European Association for Psychology and Law) and social media (Twitter), non-Dutch CPS workers were only recruited via social media pathways (Twitter). Thus, it is possible that our sample is not representative of the wider population of CPS workers. Moreover, our results are self-reports of beliefs and therefore no inferences regarding the accuracy of these can be drawn based on our data. Nevertheless, we have embedded our results regarding beliefs in the current literature in the discussion. Moreover, snowball sampling might have created a bias in recruiting people who work together and share beliefs regarding the topic of interest. Further, our CPS sample was on average descriptively a little younger than our research sample, and while 25% of our research sample had over 15 years of experience, only around 5% of our CPS sample had this much experience (See Table 2). This difference in experience could have impacted our results comparing the two different populations. Furthermore, it has to be acknowledged that our questionnaire only captures and reflects researchers’ and CPS workers’ opinions at the moment at which the data was collected. New research findings might lead to different opinions concerning disclosure patterns. Last, our sample of researchers included both child sexual abuse and memory experts, which could have contributed to the differences in opinions we found. Future research looking into this issue could benefit from asking experts which specific parts of a case they tend to give testimony on in court.

Overall, it is important to contextualise the results reported in this paper. This was a first empirical step into the opinions of relevant researchers and CPS workers on how children who are alleged victims of abuse disclose experience of CSA. As such, we covered several subtopics and sometimes (intentionally) kept our questions – and answer options – more general. This strategy allowed us to obtain responses from a relatively large sample of experts and to obtain a relatively good and broad overview of experts’ opinions on the issue of disclosure and how different disclosure patterns might affect statement validity. However, the degree of agreement with, for example, the prevalence of disclosure patterns such as false denials or recantations, might vary based on the specific context under which such phenomena could occur. Therefore, future studies might focus on possible differences of opinions regarding false denial rates among different populations of children (e.g. children who come before authorities for abuse claims versus children who are interviewed without previous disclosures).

Conclusions and Implications

Debate exists on whether expert witness reports on CSA disclosure should be brought into court (London et al., 2020; Lyon, 2002; Lyon et al., 2020). One of the arguments against admitting such evidence could be that there is no general agreement among researchers about certain elements of disclosure. Indeed, one of the Daubert criteria dictates that expert witness evidence should only be allowed regarding theories and concepts which are generally accepted in the scientific community (Daubert v. Merrell Dow Pharmaceuticals, Inc., 1993). Specifically, this is the criterium that was based on the Frye standard (Frye v. United States, 1923), which is still used as the sole guideline for admissibility for expert witness evidence in some states in the US today. Our survey results showed that there are many aspects of CSA disclosure, for instance the frequency of denials and recantations, which are disputed in the scientific – and practice oriented – community. Hence, expert witness testimony on these might violate the Daubert criteria. However, we also showed that general agreement regarding certain aspects of disclosure patterns exists: there was general agreement that (1) non-disclosure and delayed disclosure occur often, (2) false denials and recantations can happen, and (3) if they do happen, later statements can be negatively affected, but should still be admissible as evidence in court. Perhaps expert witness evidence regarding sexual abuse disclosure could best be limited to these points of general agreement.

Indeed, providing expert witness testimony on aspects with less agreement might be problematic. For example, a minority of researchers and CPS workers believed that abused children would always deny and recant their experiences. If experts who have such extreme, disputed opinions would testify regarding the frequency of false denials and recantations, then the court might reach a different conclusion compared to a case in which experts with less extreme opinions would testify. Moreover, one might wonder about interviews conducted by experts who believe that abused children always deny or recant allegations of abuse and how that might affect their line of questioning.

Our findings also have practical implications. Indeed, CPS workers agreed that the disclosure process can also include false denials and recantations. Thus, it seems unlikely CPS agents would automatically dismiss that follow-up actions might be necessary when a child denies in an initial interview. This is important because our findings also indicated that CPS workers might not be aware of how the validity of statements can be affected by the disclosure process.

We have identified areas of general agreement and dispute among experts with regards to CSA disclosure. Our results clearly show that even though certain disagreements do exist among researchers, there is also much agreement, nuancing the current debate on disclosure patterns surrounding abuse in the literature.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at , and .

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All materials and data are available on the Open Science Framework (https://osf.io/8jwtb/?view_only = 2cb15ce851464e8eb88a5b1639120ac7).

Additional information

Funding

The paper was supported by KU Leuven under a C1 grant [STG/18/011] awarded to the last author; Fonds Wetenschappelijk Onderzoek Research Project Grant [G0D3621N] awarded to the last author; and a Fonds Wetenschappelijk Onderzoek PhD fellowship Grant [11K3121N] awarded to the first author.

Notes

1 Of note, both of these approaches have certain limitations related to these biases and the populations studied (see Bücken et al., 2022).

2 Of note, we did not ask what specifically the experts’ role in these cases was and what aspect of a case they testified on.

3 Because we had not preregistered any inferential analyses, we performed a power sensitivity analyses to estimate the power we had to find such an effect. This G*Power analysis (Faul et al., 2007; t-test: difference between two independent means (two groups), α = .05, β = 0.8, group 1 N = 199, group 2 N = 267) indicated that we had enough power to find an effect sized d = 0.263 or larger.

4 These latter questions were only shown to participants when they indicated that they thought children who denied might eventually come forward with statements of abuse if interviewed suggestively or according to research based practices (answers ≠ never). Therefore, the sample size is different there.