Who Learns What? Sustained Knowledge Retention Six Months after Receiving a School Delivered Violence Awareness Programme Among Primary School Children in the UK

Abstract

This paper reports findings from a UK-wide evaluation of Speak Out Stay Safe, a manualised child abuse and neglect (CAN) prevention programme delivered by the National Society for the Prevention of Cruelty to Children (NSPCC) to children aged 6–11 years. The study examined which groups benefit most, the immediate impact on children’s knowledge of CAN, and the retention of this knowledge and help-seeking attitudes over time. Three survey measures assessed children’s understanding of abuse types, readiness to seek help, and perceptions of school culture at baseline, immediately post-intervention, and six months later. Statistical analyses included paired and independent t-tests, MANCOVAs, and Chi-square tests. Children showed significant immediate improvements in identifying neglect, sexual, emotional, and physical abuse, with knowledge largely retained at six months. Older pupils demonstrated the strongest sustained gains, particularly in recognising trusted adults and understanding sexual and emotional abuse. Girls showed significantly higher awareness and differentiation across abuse types, especially those with lower initial knowledge. Future prevention programmes should focus in depth on the concept of neglect and sexual violence in particular as these were areas where children’s knowledge was lower, especially among boys and those in more deprived school settings.

Keywords

“violence awareness”“school-based intervention”“children under 12”“help-seeking”“child abuse and neglect prevention programme”

Introduction

Childhood exposure to abuse and neglect produces both immediate harm and long-term sequelae across physical, emotional, and developmental domains (Felitti et al., 1998; Finkelhor et al., 2007; Font & Berger, 2015; Font & Maguire-Jack, 2016; Gondek et al., 2021); preventing harm and promoting timely disclosure and help-seeking in early childhood are therefore public health and safeguarding priorities. There is growing interest in the potential for children to be supported to improve their knowledge and to learn skills to help them stay safe and avoid child abuse, neglect and bullying (Stanley et al., 2021). One part of this strategy involves universal school-based programmes addressing children’s experiences of bullying and maltreatment, with programmes having been developed and delivered in schools for over 30 years (Che Yusof et al., 2022; Gubbels et al., 2021; Xu et al., 2024).

While the relevance of these programmes for adolescents is widely acknowledged (UNESCO & United Nations Entity for Gender Equality and the Empowerment of Women, 2016), debate continues about their appropriateness for younger children in primary/elementary education. Targeting of programmes has been recommended (Stanley et al., 2015), but there is little understanding about whether these programmes can boost the knowledge of those children whose awareness and knowledge are lower than those of their peers. Moreover, while some programmes might appear promising in the immediate aftermath of delivery, there is less certainty about what makes for sustained impact. The diversity of children, their cognitive development, disability status, cultural and linguistic background, prior exposure to adversity, and access to supportive adults, also appear to moderate outcomes but is often under-examined in evaluation designs (Stanley et al., 2015)

In response to these gaps, this paper examines both the immediate and longer-term effects of educational and preventive interventions over time. The study is drawn from a UK-wide evaluation (Stanley et al., 2023) of a national school-based intervention, Speak Out Stay Safe (SOSS), developed and delivered by the National Society for the Prevention of Cruelty to Children (NSPCC) in the UK. Here, we explore the gains made from this short, low-cost intervention and consider which and for whom benefits were sustained, drawing out implications for future design and targeting of school-based programmes for this age group.

Background

Children are not a homogenous group, and school-based programmes cannot assume that age or gender alone determine how children learn about abuse and neglect, or how they seek help. Children’s experiences and learning trajectories are shaped by a wider set of intersecting factors, including developmental stage, school climate, socio-economic status, faith background and geographical location (Hoefnagels et al., 2021). A programme that succeeds in one community or school setting may require adaptation to achieve the same impact elsewhere, while children’s engagement with programme content can be influenced by family culture and circumstances, peer relationships, or the resources available in local services. This complexity highlights the need for evaluations to move beyond crude demographic categories when assessing programme effectiveness, and instead to ask what truly constitutes success for different groups of children in different contexts.

While there is a lack of robust evidence on school-based programmes addressing the full spectrum of harms as covered by the programme evaluated here, some evaluations that have measured change subsequent to programme delivery have been undertaken on programmes which include more than one type of harm or abuse for this age group (Celik, 2024). For instance, Kenny et al.’s (2022) US study of the Child Safety Matters (CSM) programme with minority children in low-income areas found significant gains in knowledge over time (up to five months) regardless of age or gender. Bright et al. (2022) randomised control trial of the CSM programme indicated that increased knowledge was retained up to seven months following the programme.

A rapid evidence review conducted for this study in 2018 (Stanley et al., 2021) found no studies that covered all forms of abuse included in the SOSS programme, although there was substantial evidence on school-based child sexual abuse (CSA) and bullying programmes. Two systematic reviews of CSA prevention programmes (Fryda & Hulme, 2015; Walsh et al., 2015) indicated that children’s knowledge of CSA increased across most evaluations, although older children made better gains in knowledge than younger children when outcomes were measured using surveys, but not when vignette-based measures were used. This suggests that measures used for testing gains in knowledge need to be tailored appropriately. For example, Barron and Topping’s (2013) evaluation of the Tweenees CSA prevention programme in Scotland found that older children possessed more knowledge at the start but that younger children gained more.

Nickerson et al.’s (2019) USA study of a CSA school programme for children age four to 12 indicated significantly improved knowledge and readiness to report, with younger children making significantly greater gains and girls scoring higher than boys. Yet retention was only measured one-week post-intervention, leaving important questions about long-term effectiveness unanswered. This reflects a wider problem: programme evaluations often report short-term gains and demographic differences, but rarely consider how these play out across time or in relation to children’s broader social environments. More recent evidence further highlights that abuse prevention needs to extend beyond classroom-based programmes for children. A systematic review by Russell et al. (2024) found that prevention programmes which actively engaged parents were more likely to reinforce children’s learning, support skill retention, and facilitate protective conversations beyond the classroom.

School-based evaluations of bullying and peer violence prevention provide additional insights. Systematic reviews show that such programmes are generally effective, with stronger results from interventions lasting less than one year and for children under the age of 10 (Jiménez-Barbero et al., 2016). Implementation fidelity and contextual factors, such as teacher enthusiasm, socio-economic inequalities across schools, leadership support, and curriculum integration, are identified as crucial enablers of success, while booster sessions are recommended to sustain knowledge (Swift et al., 2017). These findings reinforce the importance of moving beyond a “one size fits all” model and towards contextually attuned approaches that account for diversity among children, schools, and communities.

The evidence indicates that while school-based programmes can improve younger children’s knowledge and readiness to seek help, a deeper understanding of diversity among children is essential to their design and evaluation. Age and gender matter, but they are only part of a larger picture in which developmental, socio-economic, cultural, institutional, and geographical factors shape how children experience and retain programme learning. This study aims to identify the immediate and longer-term gains in children’s awareness of various forms of harm, and self-reported help seeking behaviour and to establish whether children with lower levels of knowledge than their peers can benefit from receiving a child abuse and harm school-based programme. The full evaluation results are available elsewhere (Stanley et al., 2023).

Our key research questions are:

1. Can knowledge of child abuse and harm be retained by children aged 5-11years over time from a relatively brief intervention?

2. Can educational school-based programmes succeed in boosting the knowledge of the children with the lowest levels of awareness?

Speak out Stay Safe

The Speak Out Stay Safe (SOSS) programme, designed for primary school children aged 5–11 by the NSPCC in the UK, aims to enhance children’s awareness and comprehension of various forms of child abuse, neglect and bullying. It encourages them to seek assistance from a trusted adult when needed. This programme is provided free of charge to all primary schools in the UK, with the latest uptake during the 2023/24 academic year being 22.8% of schools across the UK. In that year, it reached approximately 1.4 million children across 5605 schools, including 69 Special Educational Needs and Disabilities (SEND) deliveries as reported by NSPCC (NSPCC, 2024). At the time of this evaluation, delivery of SOSS in classrooms was facilitated by NSPCC staff and trained volunteers working in pairs. The brief, low cost programme (see Stanley et al., 2021) for information on the economic evaluation of SOSS) was offered to schools regularly with the goal of ensuring every child received it twice during their primary school years. It was delivered in two versions tailored for children aged 5–7 years and 7–11 years. The programme evaluated here included a presentation delivered either to classes or in a large assembly, lasting 20 minutes for younger children and 30 minutes for older ones. Older children also participated in a one-hour interactive workshop. Utilizing child-friendly elements such as films, animation, narrative, stickers, and discussions, SOSS (now available in an online format) educates children about various forms of abuse and neglect, including physical, sexual, emotional abuse and bullying, and provides information about Childline, NSPCC’s free telephone and online helpline for children. In the programme evaluated, the assembly for older children also addressed domestic abuse. In the interactive workshop, older children delved deeper into understanding abuse and why it is essential to have someone to confide in when worried. SOSS emphasizes children’s right to speak out, feel safe, and receive assistance if necessary. Most schools received the programme once every two or three years, at nil cost to the school, with an option to engage in fundraising activities afterwards, of which approximately 40% chose to participate, with funds raised being donated to NSPCC.

Methodology

This paper focuses on those children who received the intervention. We describe the immediate improvements in knowledge and changes to reported help seeking in both the immediate (2 weeks post-delivery of SOSS) and longer-term (6 months post-delivery of SOSS) and examine differences between groups of children. While the original evaluation (Stanley et al., 2021) used a matched comparison group, this study focused on the children receiving the intervention in order to highlight shifts in their knowledge and awareness.

The survey development process drew from a review of existing tools and measures (Stanley et al., 2021) which found a lack of tested measures suitable for children under 12 covering the full spectrum of violence and abuse addressed by SOSS. Consequently, the survey integrated a bespoke tool crafted by the research team (informed by existing tools such as the What If Situations Test (WIST) (Wurtele et al., 1998), alongside validated measures. The feasibility of the methods and tools was assessed through a pilot study involving six schools (Barter et al., 2022), leading to adjustments in the survey.

To ensure variation in schools, participating sites were selected through a quota-based sampling strategy spanning all four UK nations. This approach drew on factors highlighted in the Rapid Evidence Review (Barter et al., 2022) as likely to shape programme outcomes. Three dimensions were prioritised: levels of socioeconomic disadvantage within the student body, captured through the proportion of pupils eligible for free school meals; the presence or absence of a faith designation, reflecting the school’s religious ethos; and the geographical location of the school, distinguishing between urban areas and rural or semi-rural communities.

Participants

In each school in the study, one class of younger children aged 6-8 and one class of older children aged 9-11 participated. In total, 1841 children (886 younger aged children and 955 older aged children) completed the baseline survey prior to programme delivery. At Time 2, where data collection took place 2 weeks after receiving the programme, a total of 1710 children (835 younger aged children and 875 older aged children) completed the survey. At Time 3, six months after the delivery of the programme, due to Covid-19, data collection could only be completed in 19 of the original 40 intervention schools included (see the limitations section), with a total of 803 children (410 younger aged children and 393 older aged children) completing the survey. Figure 1 shows the participant flow through this study in line with Consolidated Standards of Reporting Trials (CONSORT) guidelines (Schulz et al., 2010).

Figure 1.

Flow of Children through the Study

Procedure

Researchers conducted the survey with children during their classroom sessions. Children utilized hand-held digital tablets to complete the visually appealing survey at three different time-points: T1 (baseline pre-test), seven days prior to programme delivery; T2 (post-test), within 14 days following delivery; and T3 (follow-up), six months after the baseline assessment. Every school involved in the study received book vouchers valued at £200 as a mark of appreciation for facilitating the study.

Survey Design and Measures

The survey contained four measures, only three of which are relevant for this study. The three measures were:

1. A customized (bespoke) measure involved presenting children with a series of vignettes focused on different types of harm, followed by questions. These questions assessed children’s ability to identify whether a given scenario was harmful or not, their inclination to disclose, their capability to recognize a suitable trusted adult to confide in, and their understanding of various types of violence and abuse. The inquiry regarding different forms of violence and abuse was exclusively directed to older children. Both younger and older children’s readiness to disclose was gauged by inquiring whether they would confide in someone, while their ability to identify an appropriate trusted adult was examined by asking whom they would confide in from a list of potential individuals (eg friend, parent, classroom teacher). Two scenarios incorporated supplementary inquiries aimed at probing children’s attribution of responsibility for abuse (an example of this can be seen in the supplementary material). The reliability of the measure assessing children’s readiness to disclose remained consistent across age groups (α = 0.68 for younger children; α = 0.67 for older children), as did the measure assessing children’s capability to identify a trusted adult (α = 0.75 for younger children; α = 0.73 for older children). Additionally, children’s recollection of a helpline number (Childline) was evaluated.

2. The Children’s Knowledge of Abuse Questionnaire-Revised (CKAQ-R) (Tutty, 1995) assesses awareness of sexual abuse, with additional questions regarding bullying. The CKAQ-R comprises 24 true-false items, and an additional item concerning the acceptability of children sending a photograph of themselves in underwear was added by the research team to reflect more current experiences of being asked to send, or receiving sexual images (α = 0.88).

3. A shortened version of the Elementary Authoritative School Climate Survey (ASCS) (Cornell, 2015) was included for the older children. This comprised 11 items necessitating a binary response (yes/no), encompassing queries regarding children’s perceptions of their school environment and climate, as well as the expectation of receiving assistance from staff when encountering difficulties.

Children were provided with the option to skip or refrain from answering questions by selecting “I don’t want to answer” while progressing through the survey.

Analysis Conducted and Software

Survey data underwent thorough cleaning and scrutiny to identify any systematic missing data points. It was observed that younger children were more inclined to skip questions; nevertheless, overall, the frequency of missed questions remained within acceptable thresholds and decreased across the three time-points as children became more familiar with the survey content.

Individuals’ scores for the bespoke measure were combined across two dimensions: readiness to seek help and ability to identify an appropriate trusted adult. Additionally, scores for the CKAQ-R were combined to yield a total score of 25 (more details on the scoring and aggregation approach can be found elsewhere Stanley et al., 2021).

All children and schools were allocated identifiers, facilitating the matching of observations across the three time-points at both the school and child levels. Owing to the reduced sample sizes at T3 due to pandemic restrictions (see below), the decision was made to utilize individual children as the unit of analysis. Surveys were designed and delivered through Qualtrics XM and data was exported into SPSS for cleaning and preliminary data analysis (IBM Corp, nd. Qualtrics, 2018). R was used for the majority of our longitudinal analysis (R Core Team, 2020). Paired and clustered t-tests were subsequently employed to compare the scores of children at each of the three time-points. In addition to Multivariate Analysis of Covariance (MANCOVA) to test for differences between groups, chi-squared statistics were computed to measure the significance of the association between help seeking and attributing fault correctly among the older children.

Ethics

Our study took a hierarchical approach to eliciting consent. Consent was initially obtained from schools, followed by parents/carers using an opt-out approach, and subsequently from children who voluntarily opted into the study. In the classroom, children provided consent before completing the survey at each time-point, having been informed by their teacher about the survey’s purpose beforehand. Due to the sensitive nature of the survey questions, children were reassured that they could withdraw from the survey at any point. As the SOSS programme aims to promote children’s disclosure of abuse and harm, their involvement in the research alongside exposure to the programme was anticipated to potentially prompt disclosures. To address this, researchers developed a safeguarding procedure, resulting in the reporting of 35 instances of safeguarding disclosures or wellbeing concerns across 21 schools during the study period. Most of these disclosures/concerns were associated with older children, with half of them involving bullying (n = 13) or sibling violence (n = 5). The majority of these instances occurred during baseline data collection. Ethical approval for the study was obtained from the NSPCC Ethics Committee, as well as from ethics committees at (blinded for review).

Limitations

Although the majority of data had been gathered by March 2020, coinciding with the introduction of nationwide lockdown measures in the UK in response to the Covid-19 pandemic, subsequent data collection was significantly affected by school closures and restrictions in undertaking research. Following the reopening of schools in the UK in August–September 2020, remote data collection resumed, facilitated by schools’ electronic devices or traditional paper methods. Children were re-acquainted with the survey and guided through consent procedures via a brief video presentation recorded by the researcher who had previously visited the school and remained available remotely to assist teachers.

Sensitivity analysis was conducted to check differences in children’s responses to the survey, for pre- and post-lockdown periods. We employed a clustered comparison of means to examine any significant differences in scores between the children in the pre- and post-lockdown cohorts. With the exception of one survey item, which measured the ability to confide in a trusted adult among older children and showed a significantly higher score pre-lockdown with a small effect size (d = 0.277), no other differences were identified. As such, we have confidence that the findings across the pre- and post-lockdown cohorts at six months follow-up are comparable.

Results

In this section we will explore the sample characteristics and then present our findings under each of the research questions. We will then explore whether knowledge can be retained in the short and long term. Finally, we will present our multivariate analysis of variance for the group with the lowest scores at six months follow-up highlighting significant characteristics of this group.

The demographic characteristics of the 1841 children, with 886 aged 6–8 years and 955 aged 9–11 years, in our baseline sample can be seen in Table 1. Gender distribution was similar across age groups, with approximately half identifying as female (49.7% and 50.5%) and male (48.8% and 47.9%), while 1.6% in each group selected “I do not wish to answer” (IDWTA). As expected, age profiles aligned with the grouping criteria: the younger cohort was primarily aged 7 (55.1%) or 6 (40.4%), whereas the older cohort was largely aged 10 (47.2%) or 9 (45.3%). Participants were drawn from across the UK, with comparable proportions from England (39.2% and 40.0%), Wales (21.6% and 20.8%), Scotland (20.7% in both), and Northern Ireland (18.4% in both). Most participants resided in urban areas (69.5% and 69.0%), and the majority attended schools that reported no religious affiliation (78.3% and 75.4%). Participation in fundraising activities was reported by schools for 40.0% of 6–8-year-olds and 38.7% of 9–11-year-olds. The distribution of Free School Meal (FSM) eligibility within each school, used here as a proxy indicator for socioeconomic deprivation, was similar between cohorts, with the largest share in the most deprived quartile (Q4: 38.6% and 39.3%) and the smallest in Q3 (12.9% and 10.3%).

Table 1.

Children’s Characteristics at Baseline (T1) N = 1841

Demographic characteristics		6-8 years intervention N = 886 (%)	9-11 years intervention N = 955 (%)
Gender	Female	440 (49.7)	482 (50.5)
	Male	432 (48.8)	458 (47.9)
	IDWTA*	14 (1.6)	15 (1.6)
Age	5	2 (0.2)	0
	6	358 (40.4)	0
	7	488 (55.1)	1 (0.1)
	8	29 (3.3)	8 (0.8)
	9	0	433 (45.3)
	10	0	451 (47.2)
	11	0	28 (2.9)
	IDWTA ^a *	9 (1.0)	34 (3.6)
Country	England	348 (39.2)	382 (40.0)
	Wales	192 (21.6)	199 (20.8)
	Scotland	184 (20.7)	199 (20.8
	Northern Ireland	162 (18.4)	175 (18.4)
Geographic location	Urban	616 (69.5)	659 (69.0)
Geographic location	Rural/Semi-rural	270 (30.5)	296 (31.0)
Faith	Non-faith	694 (78.3)	720 (75.4)
Faith	Faith	192 (21.7)	235 (24.6)
Participation in fundraising activities	Non-fundraising	532 (60%)	585 (61.3%)
Participation in fundraising activities	Fundraising	354 (40%)	370 (38.7%)
%FSM quartiles^b	Q1	216 (24.4%)	247 (25.9%)
	Q2	214 (24.1%)	235 (24.6%)
	Q3	114 (12.9%)	98 (10.3%)
	Q4	342 (38.6%)	375 (39.3%)

^aIDWTA = I Don’t Want to Answer.

^bQuartiles are calculated based on national averages for children receiving this support within each of the four countries.

Descriptive statistics comparing responses at times 1 and 2 on the percentage of children who chose to tell, attribute responsibility, identify a safe adult, knew the Childline number and accurately identifying each form of abuse and harm can be seen under Tables S1-S6 in the supplementary material.

Can Knowledge of Child Abuse and Harm Be Retained by Children Aged 5-11 years Over Time From a Relatively Brief Intervention?

Knowledge and Understanding: Identifying Different Forms of Abuse and Harm

There was a statistically significant improvement in the ability of children aged 9-11 to identify neglect, sexual abuse, emotional abuse and physical abuse immediately following the delivery of SOSS, with this improvement persisting six months post intervention (Table 2). There were no significant improvements in identifying bullying which is most likely due to children’s existing higher levels of awareness at baseline. Results from the multivariate analysis indicate that children in schools with higher school climate scores (i.e. children felt the school climate was more supportive) did significantly better at identifying different types of abuse immediately after receiving SOSS, however school climate had no significant association with children’s ability to identify different types of abuse six months later.

Table 2.

Identifying different forms of abuse and harm in the short and long term Among older Children, aged 9-11

	Time 1- Time 2 (n = 798)					Time 1- Time 3 (n = 358)
Type of abuse	Mean difference	Standard deviation	z	Degrees of freedom (df)	p (2 sided)	Mean difference	Standard deviation	Z	Degrees of freedom (df)	p (2 sided)
Neglect	−0.376	0.908	−11.695	797	< .001	−0.374	0.873	−8.116	357	< .001
Sexual abuse	−0.249	0.860	−8.192	797	< .001	−0.290	0.925	−5.939	357	< .001
Emotional abuse	−0.137	0.860	−4.484	797	< .001	−0.198	0.858	−4.372	357	< .001
Bullying	0.019	0.765	0.694	797	0.488	−0.039	0.822	−0.900	357	0.369
Physical abuse	−0.076	0.747	−2.892	797	0.004	−0.117	0.769	−2.887	357	0.004

Help-Seeking: Ability to Identify an Appropriate Trusted Adult

For both age groups of children who received SOSS there were immediate and statistically significant improvements at Time 2 in their ability to identify an appropriate trusted adult to confide in. This improvement was sustained six months later for older children, but not for children aged 6-8 years. Among children aged 6-8, girls were significantly more likely to show improvements both in the short and long term. Whereas for children aged 9-11, those attending schools with higher school climate scores were significantly more likely to identify an appropriate trusted adult at both time points.

Help-Seeking: Readiness to Tell

While there was an increase in children’s readiness to tell a trusted adult immediately following receipt of SOSS, this was only significant for children aged 9-11 years. However, this improvement was not sustained six months later (Table 3).

Table 3.

Paired t-Test

Knowledge retention over time			Statistical analysis of all measures
			Readiness to tell a trusted adult			Ability to identify an appropriate trusted adult			Ability to identify and locate the childline number			Children’s ability to allocate responsibility for abuse appropriately			Children’s knowledge of sexual abuse and bullying (CKAQ-R)
			x̄ (SD)	z	P ^a	x̄ (SD)	z	P ^a	x̄ (SD)	z	P ^a	x̄ (SD)	z	P ^a	x̄ (SD)	z	P ^a
Short term 2 weeks	6-8 N = 754	T1	13.5 (2.6)	1.33	0.184	13.3 (3.0)	2.27	0.023	0.7 (3.0)	9.05	<0.001	N/A			N/A
	6-8 N = 754	T2	13.8 (2.6)	1.33	0.184	15.0 (3.0)	2.27	0.023	1.3 (3.0)	9.05	<0.001	N/A			N/A
	9-11 N = 807	T1	16.2 (3.0)	2.23	0.026	21.3 (2.6)	8.72	<0.001	1.0 (3.3)	13.1	<0.001	3.7 (2.9)	3.36	<0.001	16.6 (4.3)	2.09	0.004
	9-11 N = 807	T2	16.7 (3.0)	2.23	0.026	29.6 (2.6)	8.72	<0.001	1.7 (3.3)	13.1	<0.001	4.3 (2.9)	3.36	<0.001	18 (4.3)	2.09	0.004
6 months	6-8 N = 363	T1	13.5 (2.6)	1.75	0.080	13.3 (3.0)	0.66	0.510	0.7 (3.0)	4.85	<0.001	N/A			N/A
	6-8 N = 363	T3	14.2 (2.7)	1.75	0.080	15.3 (5.5)	0.66	0.510	1.1 (0.8)	4.85	<0.001	N/A			N/A
	9-11 N = 358	T1	16.2 (3.0)	1.52	0.13	21.3 (2.6)	4.15	<0.001	1.0 (3.3)	5.60	<0.001	3.7 (2.9)	0.06	0.531	16.6 (4.3)	1.97	0.049
	9-11 N = 358	T3	16.7 (3.1)	1.52	0.13	28.1 (13.6)	4.15	<0.001		5.60	<0.001	3.9 (2.9)	0.06	0.531	18.5 (5.2)	1.97	0.049

^aSignifies a two-sided p-value.

Help-Seeking: Ability to Identify and Locate the Childline Number

The ability of children in both age groups in receipt of SOSS to recognise and locate the Childline number was significantly improved at T2 and this was maintained at T3, with this significant improvement found for both the younger and older age groups of children (Tables 4 and 5). For the younger children, those attending non-faith schools did significantly better in the short-term and girls scored significantly better at both time points. Among the older children, those attending a faith school did significantly better in the short term, however, at both time points, older children attending schools with a higher school climate score were significantly more likely to identify and locate the Childline number.

Table 4.

Multivariate Analysis of Change Score by Measure Reporting Estimate (P-Values) for the Younger Children in the Short and Long Term

Measure	Independent variable	Reference	Younger children
			Two weeks N = 754				Six months N = 363
			Estimate	Standard error	t	p	Estimate	Standard error	t	p
BS1 readiness to tell	Sex	Girls	−0.677	0.176	−3.849	<0.001	−1.116	0.310	−3.607	<0.001
	Free school meals quartile	Q1 (lowest % of FSM)	−0.097	0.109	−0.890	0.373	−0.243	0.225	−1.080	0.280
	Location of school	Urban	−0.016	0.245	−0.064	0.949	−0.392	0.456	−0.861	0.389
	Faith	Non-faith	0.006	0.360	0.016	0.987	0.432	0.395	1.094	0.274
	Fundraising status	Fundraising	0.311	0.238	1.305	0.192	−0.252	0.463	−0.545	0.586
BS2 ability to identify an appropriate trusted adult	Sex	Girls	−1.340	0.411	−3.260	0.001	−1.853	0.708	−2.616	<0.001
	Free school meals quartile	Q1 (lowest % of FSM)	−0.048	0.243	−0.199	0.843	−0.291	0.427	−0.682	0.496
	Location of school	Urban	0.669	0.604	1.108	0.268	−1.557	0.784	−1.986	0.047
	Faith	Non-faith	−1.780	0.695	−2.561	0.010	0.528	0.607	0.870	0.384
	Fundraising status	Fundraising	0.404	0.609	0.664	0.507	−0.617	0.882	−0.699	0.484
BS5 ability to identify and locate the childline number	Sex	Girls	0.012	0.060	0.203	0.839	0.128	0.078	1.648	0.099
	Free school meals quartile	Q1 (lowest % of FSM)	0.003	0.050	0.063	0.950	0.043	0.031	1.381	0.167
	Location of school	Urban	−0.160	0.118	−1.352	0.176	−0.069	0.089	−0.779	0.436
	Faith	Non-faith	−0.075	0.113	−0.666	−0.505	−0.264	0.116	−2.270	0.023
	Fundraising status	Fundraising	0.060	0.099	0.605	0.545	0.173	0.066	2.620	< 0.001

Table 5.

Multivariate Analysis of Change Score by Measure Reporting Estimate (P-Values) for the Older Children in the Short and Long Term

Measure	Independent variable	Reference	Older children
			Two weeks N = 807				Six months N = 358
			Estimate	Standard error	t	p	Estimate	Standard error	t	p
BS1 readiness to tell	Sex	Girls	−0.311	0.129	−2.410	0.016	−0.023	0.109	3.614	0.122
	Free school meals quartile	Q1 (lowest % of FSM)	0.016	0.084	0.194	0.846	0.041	0.108	0.379	0.705
	Location of school	Urban	0.139	0.212	9,655	0.512	0.340	0.234	1.451	0.147
	Faith	Non-faith	−0.014	0.176	−0.077	0.938	0.046	0.312	0.148	0.883
	Fundraising status	Fundraising	−0.687	0.208	−3.299	0.001	−0.616	0.279	−2.281	0.023
	ASCS	Time1	0.130	0.045	2.886	0.004	0.220	0.0572	3.853	< 0.001
BS2 ability to identify an appropriate trusted adult	Sex	Girls	−1.267	0.695	−1.822	0.068	−0.077	0.065	−1.194	0.233
	Free school meals quartile	Q1 (lowest % of FSM)	−0.620	0.425	−1.458	0.145	−0.866	0.624	−1.387	0.165
	Location of school	Urban	−0.010	1.290	−0.008	0.994	−1.452	1.626	−0.893	0.372
	Faith	Non-faith	2.458	1.102	2.230	0.026	−0.347	2.147	−0.162	0.872
	Fundraising status	Fundraising	−1.186	1.128	−1.052	0.293	2.171	1.571	1.382	0.167
	ASCS	Time1	0.315	0.161	1.961	0.050	0.579	0.211	2.741	0.006
BS4 awareness of five types of harm	Sex	Girls	−0.267	0.243	−1.101	0.271	−0.024	0.016	−1.480	0.139
	Free school meals quartile	Q1 (lowest % of FSM)	−0.019	0.115	−0.164	0.869	−0.456	0.269	−1.695	0.090
	Location of school	Urban	−0.137	0.364	−0.378	0.706	−0.440	0.620	−0.709	0.478
	Faith	Non-faith	0.661	0.366	1.806	0.071	0.340	0.793	0.428	0.668
	Fundraising status	Fundraising	−0.105	0.248	−0.424	0.672	−0.105	0.595	−0.177	0.859
	ASCS	Time1	0.177	0.046	3.872	< 0.001	0.130	0.071	1.838	0.066
BS5 ability to identify and locate the childline number	Sex	Girls	−0,003	0.043	−0.077	0.939	−0.005	0.002	−2.521	0.012
	Free school meals quartile	Q1 (lowest % of FSM)	0.009	0.019	0.477	0.633	−0.094	0.038	−2.521	0.014
	Location of school	Urban	−0.010	0.043	−0.233	0.816	0.034	0.066	0.511	0.610
	Faith	Non-faith	0.065	0.055	1.200	0.230	−0.015	0.085	−0.178	0.859
	Fundraising status	Fundraising	−0.006	0.043	−0.144	0.885	0.009	0.071	0.121	0.903
	ASCS	Time1	0.016	0.008	1.960	0.050	0.027	0.013	2.085	0.037
BS6 ability to allocate responsibility for abuse appropriately	Sex	Girls	−0.285	0.175	−1.629	0.103	−0.004	0.008	−0.532	0.595
	Free school meals quartile	Q1 (lowest % of FSM)	−0.031	0.071	−0.442	0.659	0.025	0.140	0.181	0.856
	Location of school	Urban	0.047	0.168	0.282	0.778	0.151	0.327	0.463	0.643
	Faith	Non-faith	−0.234	0.146	−1.611	0.107
	Fundraising status	Fundraising	0.143	0.135	1.056	0.291	−0.556	0.247	−2.254	0.024
	ASCS	Time1	−0.013	0.024	−0.527	0.598	0.102	0.054	1.889	0.059
CKAQ-R	Sex	Girls	−0.475	0.217	−2.190	0.029	−0.018	0.013	−1.393	0.164
	Free school meals quartile	Q1 (lowest % of FSM)	−0.302	0.201	−1.506	0.132	−0.354	0.471	−0.752	0.452
	Location of school	Urban	−0.045	0.472	−0.096	0.923	−0.160	0.882	−0.181	0.856
	Faith	Non-faith	0.327	0.491	0.666	0.505	−0.720	0.991	−0.726	0.468
	Fundraising status	Fundraising	−0.095	0.370	−0.257	0.797	0.320	1.066	0.300	0.764
	ASCS	Time1	0.137	0.058	2.366	0.018	0.213	0.118	1.811	0.070

Knowledge and Understanding: Children’s Ability to Allocate Responsibility for Abuse Appropriately

Older children’s ability to allocate responsibility correctly for the harm experienced by a child was assessed via two questions in the survey. Two fault-attribution questions (for smacking and emotional abuse) made up this measure and were asked of every child aged 9-11 years, regardless of whether or not they would ‘tell’ for the associated scenario-based question. There was a statistically significant improvement in older children’s ability to allocate responsibility for abuse appropriately in the immediate aftermath of receiving SOSS. However, this improvement was not sustained over time (Table 3). Results from the multivariate analysis indicate that children in schools that took part in fundraising activities which may have increased their exposure to NSPCC material did significantly better at six months follow-up.

To test the assumption that older children’s ability to attribute fault was statistically significant among those children who recognised the scenarios above as abusive and uncomfortable (therefore opting to tell someone about it as opposed to those who would not tell or did not know), a Chi-square test was run at each time point (Table 6), the results indicated that at T1 and T2, all children who were able to attribute fault correctly were also more likely to choose to tell, with this finding being statistically significant.

Table 6.

Pearson’s Chi-Square for Fault Attribution at T2 and T3

Fault attribution	Immediate (two weeks)			Six months
Fault attribution	X²	Df	P (two-sided)	X²	Df	P (two-sided)
Whose fault is it your friend got hurt?	25.625	1	0.000	45.662	1	0.000
Whose fault is it your friend is upset?	3.763	1	0.052	23.201	1	0.000

At T3, the older children who recognised the scenario presented in the question in relation to smacking as abusive, by choosing to tell, were able to attribute fault correctly compared with their peers who chose not to tell. This finding was statistically significant. The older children who chose to seek help (tell) following the ‘what if’ question relating to emotional abuse, were also significantly more likely to attribute fault correctly, based on their willingness to seek help (tell): showing an improvement in the immediate and long-term.

Knowledge and Understanding: Sexual Abuse

We primarily assessed knowledge and understanding of child sexual abuse via the CKAQ-R measure with older children aged 9-10 years only. Fewer children chose the “I don’t want to answer” option at T2, highlighting a decreased lack of knowledge or inhibition. However, there was still a minority of children who chose the ‘wrong’ response after participating in SOSS. This was as high as 30% of the children in response to the statement: “Even someone you like could touch you in a way that feels bad”. The CKAQ-R did not show any difference by gender at baseline but did at T2, with girls scoring significantly higher on their knowledge of sexual abuse and bullying shortly after receiving the intervention. There was an overall significant improvement immediately after receiving the programme, with girls more likely than boys to score higher on this measure. This overall improvement was sustained at six months with no gender differences (Tables 3 and 5). Children’s scores between T1 and T3 showed a small effect size (d = 0.346).

Can Educational School-Based Programmes Succeed in Boosting the Knowledge of the Children With the Lowest Levels of Awareness?

Children at Baseline With Lower Levels of Knowledge and Intention to Act

We compared the proportion of children in schools in each age group who were in the bottom quartile of scores for each of the measures at baseline, and whether they remained in this quartile at six months follow-up. The demographic characteristics of the younger and older children with lower levels of knowledge and intention to act show enduring trends across age groups. Lower scores in younger children (n = 689) were significantly associated with being male, eligible for FSMs, and attending urban schools. These younger children were less likely to report readiness to tell, identify or confide in a trusted adult, or know how to contact Childline. For older children (n = 676), similar associations emerged, whereby boys and those receiving FSMs consistently scored lower across measures of readiness to tell, identifying and confiding in a trusted adult, and correctly allocating responsibility for abuse. In addition, faith school attendance and non-participation in fundraising were linked to reduced awareness of the Childline number (Stanley et al., 2021). The findings from the older children further indicate that socioeconomic disadvantage and gender disparities extend into later primary years, with implications for targeted safeguarding education and equitable delivery of school-based programmes.

The low at baseline group showed improvements across a range of help-seeking and knowledge measures in the subsequent six months after the receipt of SOSS (Tables 7 and 8). Children in the lowest quartile of scores at baseline all scored zero on their knowledge of the Childline number and where to find it.

Table 7.

Multivariate Analysis of Change in Scores in the Short and Long Term for Younger Children Scoring in the Lowest Quartile at T1 by Measure and Demographic Characteristics Reporting Estimate* N = 689

Measure	Time period	Free school meals (lowest quartile as reference)			Faith schools	Gender (Boys)	Urban school setting	Schools not participating in fundraising
Measure	Time period	Q2	Q3	Q4	Faith schools	Gender (Boys)	Urban school setting	Schools not participating in fundraising
BS1 readiness to tell	Two weeks	0.511	1.467	−0.138	−0.364	−1.062*	0.740	0.44
BS1 readiness to tell	Six months	−2.096*	−4.063**	−2.922*	1.588	−0.737	−1.040	−0.719
BS2 who to tell	Two weeks	2.111	−1.615	0.951	−2.421	−0.694	−1.185	2.038
BS2 who to tell	Six months	−3.395**	−4.617	−5.036**	0.748	−1.582	−2.289	0.571
BS5 childline	Two weeks	0.187	0.084	0.175	−0.204	0.048	−0.307	−0.059
BS5 childline	Six months	−0.227		0.220	0.180***	0.082	−0.088	0.103

*≤ 0.05 ** ≤0.01 *** ≤0.001.

Table 8.

Multivariate Analysis of Change in Scores in the Short and Long Term for Older Children Scoring in the Lowest Quartile at T1 by Measure and Demographic Characteristics Reporting Estimate* N = 676

Measure	Time period	Free school meals (lowest quartile as reference)			Faith schools	Gender (Boys)	Urban school setting	Schools not participating in fundraising	School climate at baseline
Measure	Time period	Q2	Q3	Q4	Faith schools	Gender (Boys)	Urban school setting	Schools not participating in fundraising	School climate at baseline
BS1 readiness to tell	Two weeks	−0.751	−0.826	0.053	1.243**	−1.586***	−0.188	−2.052***	0.177
BS1 readiness to tell	Six months	−0.185	−0.604	−0.849	0.589	0.717*	0.628	−2.474*	0.349**
BS2 who to tell	Two weeks	−5.476**	0.435	−1.942	5.217**	−2.737	−3.678	−1.791	0.473**
BS2 who to tell	Six months	−0.584	4.674	−2.456	0.972	−5.623	−3.655	0.428	0.562
BS4 recognition of different harms	Two weeks	−1.065	−1.596**	−0.226	0.560	0.303	−0.361	−0.286	0.191**
BS4 recognition of different harms	Six months	−1.916	−0.018	−2.474*	2.088	−2.600**	−1.467	0.014	0.079
BS5 childline	Two weeks	−0.194	0.132	0.085	0.173	0.012	−0.007	−0.103	0.007
BS5 childline	Six months	−0.194	0.237	−0.400	−0.001	−0.491**	0.120	−0.124	0.021
BS6 allocation of responsibility	Two weeks	−0.556	−0.423	−0.174	0.343	−0.606	−0.257	−0.096	−0.035
BS6 allocation of responsibility	Six months	−0.786	−0.304	−1.148	1.107	−0.552	0.717	0.241	0.163
CKAQ-R sexual abuse & bullying	Two weeks	1.713	0.316	−1.474	1.582	−1.133	0.754	−1.236*	0.183*
CKAQ-R sexual abuse & bullying	Six months	0.347	−0.688	−1.662	0.657	−1.554	−0.310	2.039	0.103

*≤ 0.05 ** ≤0.01 *** ≤0.001.

The most notable change was in the proportion of low-scoring children who could identify an appropriate trusted adult to tell, or had knowledge of the Childline number and where to find it: nearly a third of children aged 6-8 and a quarter of children aged 9-11 in the bottom quartile shifted into a higher quartile and sustained this improvement over time. Among the older low-scoring children (aged 9-11), boys scored significantly lower than their female peers at six months follow-up in their ability to recognise different forms of harm and identify and locate the Childline number despite no gender difference in the short term.

Some children aged 6-8 years who were low at baseline experienced statistically significant improvements in all dimensions of help-seeking between baseline and six months follow-up (p < 0.01). There are indications that SOSS may be responsible for improving some children’s ability to confide in a trusted adult, as SOSS aims to help some of those children who are the most reluctant to speak out to a trusted adult, by boosting their willingness to seek help. Similarly, some low scoring children aged 9-11 years also made statistically significant improvements between baseline and six months follow-up in all dimensions of help-seeking (p < 0.01).

Improvements in knowledge levels were sustained in children with low baseline scores over the six months. While children’s ability to correctly allocate responsibility for abuse (BS6) did not change, knowledge of sexual abuse and bullying improved significantly between baseline and immediately after receiving the intervention and at six months follow-up (CKAQ-R, p < 0.01). Short term improvement was evident in schools with higher school climate scores and among schools who participated in fundraising. Awareness of different types of abuse or harm also improved significantly (p < 0.01), particularly among low-scorers who were female, attended faith schools, or attended schools in the highest quartile of FSM. However, for boys in the lowest quartile of scores at baseline, they scored significantly lower than their female peers at six months follow-up.

Socioeconomic status (represented by free school meals) was consistently associated with knowledge retention in the long term for children scoring in the lowest quartile at baseline, particularly in their readiness to tell and understanding of “who to tell.” However, there are limited significant predictors for short-term changes. School type, faith school and urban setting, showed significant impacts on disclosure readiness and harm recognition, especially in the short term. School climate also emerges as a significant factor influencing disclosure readiness over time.

Discussion

The present study set out to address two critical research questions. First, can CAN knowledge be retained over time following a relatively brief intervention? Second, can such programmes succeed in improving the awareness of the minority of children whose baseline knowledge is notably lower than that of their peers? These questions are of practical and policy significance in the context of CAN awareness and disclosure, where schools play a central role in equipping children with the skills to recognise maltreatment and seek help.

Our findings indicate that a brief, school-based CAN intervention can produce sustained improvements in children’s knowledge of different abuse types and in their reported help-seeking readiness. Moreover, the intervention not only benefits the general student population but also supports those starting with the lowest baseline awareness, demonstrating its potential to reduce safeguarding knowledge disparities within school cohorts.

Among older children (9–11 years), awareness of neglect, sexual, emotional, and physical abuse significantly increased and these gains persisted at six months; consistent with global evidence indicating that school-based awareness programmes can enhance knowledge and protective behaviours, with retention over time (Che Yusof et al., 2022; Hoefnagels et al., 2021; McElearney et al., 2021). We have extended this evidence base by focusing on primary/elementary school aged children across the four nations of the UK. A positive school climate was associated with stronger immediate knowledge gains, although gains in ability to identify different forms of harm and abuse were not sustained at six months. However, improved ability to identify a trusted adult and increased knowledge of Childline were retained at six months by older children in our sample and these gains were significantly associated with higher school climate scores. This mixed picture aligns with mounting evidence that a supportive school environment underpins mental well-being and resilience, yet sustained improvement often requires ongoing reinforcement (Franco et al., 2022; Hansen & Barene, 2025; Leurent et al., 2021). Both age groups showed improvement in identifying trusted adults. However, sustained improvement was observed only among older children and younger girls, underscoring the importance of gender and age-responsive tailoring in safeguarding interventions, a finding echoed in broader literature on disclosure and self-efficacy (Che Yusof et al., 2022; Lu et al., 2023).

Older children showed initial improvement in attributing fault correctly, although these gains faded by six months, except in schools involved in NSPCC-linked fundraising activities, again suggesting that ongoing exposure and reinforcement enhance long-term conceptual understanding. This supports literature advocating for multi-component, whole-school designs to bolster retention (Lu et al., 2023; Podiya et al., 2025).

Importantly, children initially in the lowest scoring quartile made meaningful improvements, particularly in help-seeking competencies. Yet older boys in this group continued to lag behind girls at follow-up. This underscores the necessity for targeted strategies to close gendered safeguarding disparities, echoing findings that suggest socio-cognitive mechanisms and gender norms mediate engagement with CSA education (Che Yusof et al., 2022). Socioeconomic status also appeared to mediate gains for this group and this finding supports arguments for targeting intervention on areas where social and economic needs are high. School climate appeared to mediate immediate gains for this group; however, baseline measures of school climate were not consistently associated with sustained improvement over time. An exception emerged for readiness to tell, particularly among girls and those attending faith schools or participating in fundraising activities, where a more positive school climate seemed to support continued progress. Help-seeking behaviour, conceptualised here as readiness to tell, should not be viewed as a single event but rather as a dynamic, relational process that develops over time (Reitsema & Grietens, 2016). Each interaction reshapes the meaning of help-seeking, with the responses and characteristics of those receiving the request being just as pivotal as the child’s own words or actions (Hoefnagels et al., 2021; McElvaney et al., 2012). This underscores the importance of context and interactional dynamics in influencing whether, when, and how children seek help. Because school climate is dynamic rather than fixed, our findings suggest robust measurement requires repeated assessment over extended periods to capture change and continuity.

This study deliberately focuses on school-based awareness-raising programmes, which offer the advantage of broad reach across child populations but have also been critiqued for implicitly placing the onus of protection on children themselves by assuming that the key protective action is for children to “speak out.” In response to this critique, emerging evidence highlights the value of whole-school approaches that embed prevention across multiple domains of school life: programmes that integrate messages across school-wide policies, curriculum, staff development and engagement with families are posited can strengthen protective environments and support appropriate responses when children disclose harm (Gubbels et al., 2021; Russell et al., 2024). A notable example is the Keeping Safe whole-school child maltreatment prevention programme, which explicitly includes components aimed at children, parents and school staff to address maltreatment comprehensively rather than solely focusing on individual child disclosure (McElearney et al., 2018, 2021).

Under the Ofsted Education Inspection Framework in England, safeguarding and personal development remain critical inspection areas and pupils’ ability to identify trusted adults and routes for help is emphasised (Ofsted, 2023). The outcomes of SOSS are directly aligned with these skills. Calls for mandatory reporting by professionals, including teachers, further underscore the need for competent pupil education on abuse and help-seeking (Ferguson, 2025; Jay et al., 2022). Moreover, the shift toward earlier, developmentally appropriate teaching on healthy relationships and consent, highlighted by primary school abuse testimonies (Everyone’s Invited, 2024; Ferguson, 2025), reinforces the importance of embedding safeguarding across the curriculum.

Recommendations for Future Practice and Research

1. Schools should embed reinforcement mechanisms such as booster sessions, parental and staff engagement, and curriculum integration to sustain knowledge and behaviour gains over time.

2. Tailor interventions for boys whose knowledge and understanding remains low, to reduce inequity in safeguarding outcomes.

3. Improve teacher and school staff training on neglect and sexual abuse, developing more resonant, developmentally appropriate messaging to support deeper understanding.

4. Strengthen school climate efforts, enhancing relational, safety, and trust-building components, which support both immediate and long-term intervention effectiveness.

5. Extend evaluation follow-up periods beyond six months to assess whether knowledge and help-seeking behaviour translate into actual protective actions.

6. Longitudinal research that explores whether knowledge and understanding acquired before 12 can protect children into adolescence would be valuable.

7. Evaluate policy integration and cost-effectiveness, exploring system-wide implementation within statutory structures to scale safeguarding impact affordably and sustainably.

Conclusion

A brief intervention like SOSS can foster improvements sustained over the medium term in children’s safeguarding knowledge and practical help-seeking capacity, aligned with UK policy goals and inspection standards. To maximise impact and equity, incorporated strategies must reinforce learning, tailor content to under-served subgroups, and integrate interventions into the broader school safeguarding infrastructure so that messages are reinforced and change is sustained.

Supplemental Material

Supplemental material - Who Learns What? Sustained Knowledge Retention Six Months After Receiving a School Delivered Violence Awareness Programme Among Primary School Children in the UK

Supplemental material for Who Learns What? Sustained Knowledge Retention Six Months After Receiving a School Delivered Violence Awareness Programme Among Primary School Children in the UK by Zain Kurdi, Ugur Ozdemir, John Devaney, Nicky Stanley in Child Maltreatment.

Footnotes

Acknowledgments

Thanks to all the pupils in England, Wales, Scotland and Northern Ireland who took part in our evaluation, their teachers and school staff who actively engaged with us while undertaking the evaluation. Colleagues from the wider research team.

ORCID iDs

Zain Kurdi

Ugur Ozdemir

John Devaney

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This independent study was funded by the National Society for Prevention of Cruelty to Children (NSPCC), the JMCMRJ Sorrell Foundation and the Health Foundation.

Declaration of Conflicting Interests

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

Data Availability Statement

Data will be made available on request.*

Supplemental Material

Supplemental material for this article is available online.

References

Barron

I. G.

Topping

K. J.

(2013). Exploratory evaluation of a school-based child sexual abuse prevention program. Journal of Child Sexual Abuse, 22(8), 931–948. https://doi.org/10.1080/10538712.2013.841788

Barter

Batool

Charles

Devaney

Farrelly

Hayes

Kurdi

Millar

Monks

Richardson Foster

Radford

Tudor Edwards

Winrow

Stanley

(2022). Conducting large-scale mixed-method research on harm and abuse prevention with children under 12: Learning from a UK feasibility study. Children & Society, 38(1), 79–96. online(n/a). https://doi.org/10.1111/chso.12658

Bright

M. A.

Sayedul Huq

Patel

Miller

M. D.

Finkelhor

(2022). Child safety matters: Randomized control trial of a school-based, child victimization prevention curriculum. Journal of Interpersonal Violence, 37(1–2), 538–556. https://doi.org/10.1177/0886260520909185

Celik

(2024). The effectiveness of school-based child sexual abuse prevention programmes among primary school-aged children: A systematic review. International Journal of Educational Research Open, 7, Article 100348. https://doi.org/10.1016/j.ijedro.2024.100348

Che Yusof

Norhayati

M. N.

Mohd Azman

(2022). Effectiveness of school-based child sexual abuse intervention among school children in the new millennium era: Systematic review and meta-analyses. Frontiers in Public Health, 10, Article 909254. https://doi.org/10.3389/fpubh.2022.909254

Cornell

(2015). The Authoritative School Climate Survey and The School Climate Bullying Survey: Research Summary (76).

Everyone’s Invited . (2024). Everyone’s invited AnnualReport 23-24. https://www.everyonesinvited.uk/EIAnnualReport23-24.pdf

Felitti

V. J.

Anda

R. F.

Nordenberg

Williamson

D. F.

Spitz

A. M.

Edwards

Koss

M. P.

Marks

J. S.

(1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The adverse childhood experiences (ACE) study. American Journal of Preventive Medicine, 14(4), 245–258. https://doi.org/10.1016/s0749-3797(98)00017-8

Ferguson

(2025). https://www.theguardian.com/world/2025/mar/22/raped-at-the-age-of-10-accounts-of-sexual-abuse-and-harassment-at-1664-uk-primary-schools-reported ‘I was raped at the age of 10’: Sexual abuse and harassment reported at 1,664 UK primary schools. The Guardian.

10.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Re-victimization patterns in a national longitudinal sample of children and youth. Child Abuse & Neglect, 31(5), 479–502. https://doi.org/10.1016/j.chiabu.2006.03.012

11.

Font

S. A.

Berger

L. M.

(2015). Child maltreatment and children’s developmental trajectories in early to middle childhood. Child Development, 86(2), 536–556. https://doi.org/10.1111/cdev.12322

12.

Font

S. A.

Maguire-Jack

(2016). Pathways from childhood abuse and other adversities to adult health risks: The role of adult socioeconomic conditions. Child Abuse & Neglect, 51, 390–399. https://doi.org/10.1016/j.chiabu.2015.05.013

13.

Franco

Baumler

Torres

E. D.

Wood

Temple

J. R.

(2022). The link between school climate and mental health among an ethnically diverse sample of middle school youth. Current Psychology, 1–11. https://doi.org/10.1007/s12144-022-03016-y

14.

Fryda

C. M.

Hulme

P. A.

(2015). School-based childhood sexual abuse prevention programs: An integrative review. The Journal of School Nursing, 31(3), 167–182. https://doi.org/10.1177/1059840514544125

15.

Gondek

Patalay

Lacey

R. E.

(2021). Adverse childhood experiences and multiple mental health outcomes through adulthood: A prospective birth cohort study. SSM - Mental Health, 1, Article 100013. https://doi.org/10.1016/j.ssmmh.2021.100013

16.

Gubbels

van der Put

C. E.

Stams

G.-J. J. M.

Assink

(2021). Effective components of school-based prevention programs for child abuse: A meta-analytic review. Clinical Child and Family Psychology Review, 24(3), 553–578. https://doi.org/10.1007/s10567-021-00353-5

17.

Hansen

K. G.

Barene

(2025). Exploring the associations between school climate and mental wellbeing: Insights from the MOVE12 pilot study in Norwegian secondary schools. European Journal of Investigation in Health Psychology and Education, 15(4), 46. https://doi.org/10.3390/ejihpe15040046

18.

Hoefnagels

Onrust

van Rooijen

Jonkman

van Spanje-Hennes

Breeman

L. D.

(2021). Changing the classroom climate to lower the threshold for child abuse and neglect self-disclosure: A non-randomized cluster controlled trial. Children and Youth Services Review, 130(1), 106196. https://doi.org/10.1016/j.childyouth.2021.106196

19.

IBM Corp . (n.d.). SPSS for windows (version version 26). Author. [Computer software].

20.

Jay

Evans

Frank

Sharpling

(2022). The report of the independent inquiry into child sexual abuse. https://www.iicsa.org.uk/reports-recommendations/publications/inquiry/final-report.html

21.

Jiménez-Barbero

J. A.

Ruiz-Hernández

J. A.

Llor-Zaragoza

Pérez-García

Llor-Esteban

(2016). Effectiveness of anti-bullying school programs: A meta-analysis. Children and Youth Services Review, 61, 165–175. https://doi.org/10.1016/j.childyouth.2015.12.015

22.

Kenny

M. C.

Long

Billings

Malik

(2022). School-based abuse prevention programming: Implementation of child safety matters with minority youth. Child Abuse Review, 31(3), Article e2742. https://doi.org/10.1002/car.2742

23.

Leurent

Dodd

Allen

Viner

Scott

Bonell

(2021). Is positive school climate associated with better adolescent mental health? Longitudinal study of young people in England. SSM - Mental Health, 1, Article 100033. https://doi.org/10.1016/j.ssmmh.2021.100033

24.

Barlow

Meinck

Neelakantan

(2023). Unpacking school-based child sexual abuse prevention programs: A realist review. Trauma, Violence, & Abuse, 24(4), 2067–2081. https://doi.org/10.1177/15248380221082153

25.

McElearney

Adamson

Murphy

Rogers

Radcliffe

(2021). Keeping safe: An effective whole-school programme that teaches children to recognise abusive behaviours and tell: cluster randomised trial evaluation results in primary schools in northern Ireland. https://learning.nspcc.org.uk/research-resources/2021/keeping-safe-cluster-randomised-trial-evaluation

26.

McElearney

Brennan-Wilson

Murphy

Stephenson

Bunting

(2018). Cluster randomised controlled trial of ‘whole school’ child maltreatment prevention programme in primary schools in northern Ireland: Study protocol for keeping safe. BMC Public Health, 18(1), 590. https://doi.org/10.1186/s12889-018-5492-8

27.

McElvaney

Greene

Hogan

(2012). Containing the secret of child sexual abuse. Journal of Interpersonal Violence, 27(6), 1155–1175. https://doi.org/10.1177/0886260511424503

28.

Nickerson

A. B.

Tulledge

Manges

Kesselring

Parks

Livingston

J. A.

Dudley

(2019). Randomized controlled trial of the child protection unit: Grade and gender as moderators of CSA prevention concepts in elementary students. Child Abuse & Neglect, 96, Article 104101. https://doi.org/10.1016/j.chiabu.2019.104101

29.

NSPCC . (2024). Author Speak out Stay safe UK wide in 2023/24. Knowledge and Information Services – FIT.

30.

Ofsted . (2023). Education inspection framework guidance. GOV.UK. https://www.gov.uk/government/publications/education-inspection-framework/education-inspection-framework-for-september-2023

31.

Podiya

J. K.

Navaneetham

Bhola

(2025). Influences of school climate on emotional health and academic achievement of school-going adolescents in India: A systematic review. BMC Public Health, 25(1), 54. https://doi.org/10.1186/s12889-024-21268-0

32.

Qualtrics . (2018). Qualtrics XM. Qualtrics. [Computer software].

33.

R Core Team . (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing. [Computer software]. https://www.R-project.org/

34.

Reitsema

A. M.

Grietens

(2016). Is anybody listening? The literature on the dialogical process of child sexual abuse disclosure reviewed. Trauma, Violence, & Abuse, 17(3), 330–340. https://doi.org/10.1177/1524838015584368

35.

Russell

D. H.

Trew

Harris

Dickson

Walsh

Higgins

D. J.

Smith

(2024). Engaging parents in child-focused child sexual abuse prevention education strategies: A systematic review. Trauma, Violence, & Abuse, 25(4), 3082–3098. https://doi.org/10.1177/15248380241235895

36.

Schulz

K. F.

Altman

D. G.

Moher

CONSORT Group . (2010). CONSORT 2010 statement: Updated guidelines for reporting parallel group randomised trials. BMJ, 340, c332. https://doi.org/10.1136/bmj.c332

37.

Stanley

Barter

Batool

Charles

Devaney

Tudor Edwards

Farrelly

Hayes

Kasperkiewicz

Kelly

Kurdi

Millar

Monks

Ozdemir

Radford

Thompson

Winrow

(2021). Evaluation of the NSPCC speak out stay safe programme: Final report. https://learning.nspcc.org.uk/media/2690/evaluation-nspcc-speak-out-stay-safe-programme-final-report.pdf.88.

38.

Stanley

Barter

Batool

Charles

Devaney

Tudor Edwards

Farrelly

Hayes

Kasperkiewicz

Kurdi

Millar

Monks

Ozdemir

Radford

others , et al. (2021). Evaluation of the NSPCC speak out stay safe programme: Appendices. NSPCC. https://learning.nspcc.org.uk/media/2693/evaluation-nspcc-speak-out-stay-safe-programme-appendices.pdf.

39.

Stanley

Devaney

Kurdi

Ozdemir

Barter

Monks

Tudor Edwards

Batool

Charles

Farrelly

Hayes

Millar

Thompson

Winrow

Radford

(2023). What makes for effectiveness when starting early – Learning from an integrated school-based violence and abuse prevention programme for children under 12. Child Abuse & Neglect, (139). https://doi.org/10.1016/j.chiabu.2023.106109

40.

Stanley

Ellis

Farrelly

Hollinghurst

Downe

(2015). Preventing domestic abuse for children and young people: A review of school-based interventions. Children and Youth Services Review, 59, 120–131. https://doi.org/10.1016/j.childyouth.2015.10.018

41.

Swift

L. E.

Hubbard

J. A.

Bookhout

M. K.

Grassetti

S. N.

Smith

M. A.

Morrow

M. T.

(2017). Teacher factors contributing to dosage of the KiVa anti-bullying program. Journal of School Psychology, 65, 102–115. https://doi.org/10.1016/j.jsp.2017.07.005

42.

Tutty

(1995). The revised Children’s knowledge of abuse questionnaire: Development of a measure of children’s understanding of sexual abuse prevention concepts. Social Work Research, 19(2), 112–120. https://doi.org/10.1093/swr/19.2.112

43.

UNESCO & United Nations Entity for Gender Equality and the Empowerment of Women . (2016). Global guidance on addressing school-related gender-based violence. UNESCO. https://doi.org/10.54675/VFOP6248

44.

Walsh

Zwi

Woolfenden

Shlonsky

(2015). School‐based education programmes for the prevention of child sexual abuse. Cochrane Database of Systematic Reviews, 2015(4). https://doi.org/10.1002/14651858.CD004380.pub3

45.

Wurtele

S. K.

Hughes

Owens

J. S.

(1998). An examination of the reliability of the “What If” situations test: A brief report. Journal of Child Sexual Abuse, 7(1), 41–52. https://doi.org/10.1300/J070v07n01_03

46.

Yang

(2024). Enhancing child sexual abuse prevention knowledge with an educational toolkit -Evaluation of the Chinese doll program. International Journal of Public Health, 69, Article 1606641. https://doi.org/10.3389/ijph.2024.1606641

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB