The Potential for Anti-Stigma Interventions to Change Public Attitudes Toward Minor-Attracted Persons: A Replication and Extension of Jara and Jeglic’s Study

ABSTRACT

From a public health perspective, ensuring access to, and uptake of, support services for people concerned about their sexual thoughts and behavior is essential to the prevention of child sexual abuse. However, public and fiscal support for these services can be adversely affected by negative preconceptions regarding minor-attracted persons (MAPs); negative stigma may also limit MAPs’ engagement with such services. Using a randomized-control design, the present study replicated and extended a recent US study to test effects of different modes of educational messaging to reduce negative attitudes toward MAPs in Australia. Participants were recruited nationally via a Qualtrics XM online survey platform. The final sample (n = 178) were aged 18–84 (M = 47.57 years, SD = 17.34; 49% male). Participants completed the Attitudes Toward Minor Attracted Persons (ATMAP) scale prior to being randomly assigned to one of four conditions (written text [facts only]; written text [myth and facts]; info-video [myths and facts]; and control). Following intervention (or control) the ATMAP was readministered; those in experimental groups also completed a comprehension test. Mixed between-within subjects analysis of covariance revealed significant reductions in pre-post scores on the ATMAP scale, but no significant differences were found across intervention types, or the control. Scores indicated that older participants were less negative in their attitudes compared to younger participants, both before and after intervention. Findings suggest that educating the public may improve attitudes toward MAPs, but more knowledge is required on how to best disseminate messages to maximize impact.

KEYWORDS:

• Minor-attraction
• anti-stigma
• prevention
• early intervention
• sexual abuse
• public attitudes

Introduction

Child sexual abuse (CSA) is a cause of community concern around the world triggering substantial emotional reaction from the general public. These concerns are not unfounded. Certainly, available prevalence rates show a significant proportion of children are affected by offline and online sexual abuse globally (e.g., Barth et al., 2012; Finkelhor et al., 2014, 2022; Moody et al., 2018; Stoltenborgh et al., 2011, 2014). In Australia, where this research is based, a recent longitudinal study (Townsend et al., 2022) indicated a CSA prevalence rate between 12% and 15% among Australian women; the Personal Safety Survey also identified that an estimated 1.4 million (7.7%) Australian adults have experienced CSA (Australian Bureau of Statistics, 2019). Given these rates and the known, and oftentimes comorbid, short- and long-term harms associated with CSA (Hailes et al., 2019; Lewis et al., 2016; Nagtegaal & Boonmann, 2021; Papalia et al., 2017), preventing CSA before it first occurs is key to reduce the extent and impacts of CSA in the community. Furthermore, with lifetime economic costs of CSA estimated to be ~$US9.3 billion (Letourneau et al., 2018), investing in initiatives that seek to intervene early (also known as upstream initiatives) is necessary to form part of an effective and comprehensive CSA prevention agenda, consistent with a public health approach to sexual violence prevention (Letourneau et al., 2014; McKillop, 2019).

Situated within a public health prevention approach, part of early intervention includes a targeted response to individuals known to be vulnerable to or at higher risk for CSA perpetration (Cant et al., 2022). One group of at-risk individuals are those who have a sexual interest in minors (known as “Minor Attracted Persons” or MAPs). Minor-attraction may be present in as many as 4% or 5% of adult men to varying degrees (Levenson & Grady, 2019a see also Sorrentino & Abramowitz, 2021). Although research links pedophilia (as one form of minor attraction) as a motivator for CSA (Seto, 2019) and to recurrent sexual offending behavior (Stephens et al., 2017), minor attraction toward children does not necessarily lead to CSA. Many MAPs (some of whom also identify as virtuous pedophiles) report successfully abstaining from offending (Cantor & McPhail, 2016). Many, in fact, actively seek out support to help manage (or suppress) their thoughts and attraction (Malone, 2016), so as not to offend.

Certainly, having support services and resources available for those with sexual interest in minors (but who are yet to offend) to assist them in abstaining from perpetrating CSA, seems a logical preventative action in which to invest. Indeed, for those who do perpetrate CSA, research has found substantial time lags between the onset of the sexual interests (such as minor-attraction) and first arrest (e.g., Piché et al., 2016). Hence, as Parr and Pearson aptly state “it begs the question of whether there is a window of opportunity between realizing one’s attraction and acting upon it” (p. 946). By inference, a lack of investment in early intervention initiatives has likely contributed to many missed opportunities to prevent the onset of CSA.

In their recent study, Jara and Jeglic (2021) make some important points regarding the need for early intervention and preventative responses to CSA, and the barriers that have impeded access to such support. Drawing on recent studies, they highlight barriers associated with: (1) community misconceptions; (2) stigma; (3) lack of access to support services; and, (4) concerns by MAPs to seek appropriate support. The authors make a compelling argument regarding how these factors, together, create roadblocks to successful implementation of upstream interventions to forestall CSA behavior among those potentially most at risk. Indeed, as Lievesley and Harper (2022) point out, such barriers limit the potential impact (and reach) that support services can have to reduce incidences of CSA and its impacts in the community.

Importantly, Jara and Jeglic (2021) discuss how barriers to early intervention are fueled by a common misconception that all individuals who commit CSA are pedophiles, and that this behavior is frequent and predatory; and thus, identify the need to correct these misconceptions through community awareness and education. Part of the barrier to community investment in early intervention is the perception that minor attraction is immutable, and MAPs are unable to control their thoughts and behavior, making them more likely to engage in CSA and to reoffend (Jahnke & Hoyer, 2013; Levenson et al., 2007). Subsequent stigma and negative rhetoric often leads to a focus on chastising MAPs, rather than seeing the potential value in supporting them (Richards, 2018). What this misconception also fails to recognize is that many MAPs are indeed concerned by their thoughts about children and also recognize the gravity of acting on these thoughts (Blagden et al., 2018; Malone, 2016).

The stigmatization of minor-attraction and MAPs involves the labeling of these individuals as “other,” with negative stereotyping, devaluation, and the active separation of MAPs from the general public (Grady et al., 2019; Lievesley et al., 2020). It has been reported that the public harbor feelings of anger and disgust toward MAPs (Levenson & Grady, 2019b; Richards, 2018); and that these individuals may experience additional difficulties in the face of substantial negative attitudes toward them (Richards, 2018). As Grady et al. (2019) acknowledge, these stigmatizing attitudes are clearly communicated by the public to MAPs, contributing to the internalization and self-stigmatization by these individuals. This self-stigmatization leads MAPs to self-isolate, fearing discovery, with detrimental impacts on self-esteem and cognitive and emotional well-being, creating personal challenges (e.g., social connections and intimate relationships) (Cohen et al., 2018; Grady et al., 2019; Lievesley et al., 2020). These effects are identified as possible risk markers that may lead to a greater potential for offending behavior and a decreased willingness to seek professional help (Grady et al., 2019; Jara & Jeglic, 2021; Levenson & Grady, 2019b; Lievesley et al., 2020; Parr & Pearson, 2019).

It has been suggested that the professional treatment and support of MAPs may decrease the influence and number of risk factors that contribute to future CSA behaviors (Jara & Jeglic, 2021), yet there remains a lack of available support services in the community when MAPs do reach out. This lack of support services is compounded by practitioners’ lack of willingness (and/or perceived comfort and competency) to work with MAPs in this capacity, which reduces workforce capacity to manage demand (Lievesley & Harper, 2022; Sorrentino & Abramowitz, 2021).

Helping-seeking is also circumvented by fears by MAPs of reprisal/retaliation (Malone, 2016; Ricciardelli & Moir, 2013). Parr and Pearson (2019) reported that majority of MAPs want professional mental health support, but that the internalized stigma prevents them from seeking help. Certainly, Grady et al. (2019) study of MAPs, among others (e.g., Levenson et al., 2017; Lievesley et al., 2020), have demonstrated how stigma creates barriers for help-seeking by MAPs. Although many MAPs suppress their attraction and/or urges to act on their sexual desires (Jara & Jeglic, 2021; Levenson et al., 2017), Lievesley et al. (2020) caution that such suppression – without professional treatment and appropriate supports – may increase the likelihood of future CSA behaviors. Certainly, Elchuk et al. (2022) found that, while disclosure itself did not reduce psychological distress for MAPs, psychological and emotional wellbeing was improved when disclosure was met with support, highlighting the potential value in engaging with support networks and services to improve outcomes. Combined, it seems that dispelling the stigma associated with minor attraction toward children and MAPs and to identify ways to break down barriers to providing (and seeking) support, is an essential component of current (and future) preventative action.

The Australian context

In Australia, secondary (or early intervention) prevention efforts have been somewhat limited, compared to other Western countries, but have gained momentum in the last five years in response to the Royal Commission into Institutional Responses to Child Sexual Abuse (2017) and the subsequent commitment by the Australian Government to introduce a national strategy to prevent child sexual abuse. As a result, prevention and early intervention initiatives are on the incline. This presents opportunities to innovate and scale up secondary prevention efforts nationally.

Early intervention is a key mechanism for preventing CSA onset. Perpetration-focused support initiatives are one component of early intervention but can be more difficult to garner community and fiscal support for (McKillop & Rayment McHugh, 2022). As already noted, acceptance of prevention and early intervention (and thus support for initiatives) is largely driven by community sentiment (and thus political appetite). Hence, introducing any new early intervention initiative requires careful planning to ensure successful uptake and sustained implementation, as has been recognized in Australia (Cant et al., 2022; McKibbin et al., 2022) and globally (Kewley et al., 2021). To plan for the successful implementation of such early intervention efforts, it is therefore important to examine community knowledge and sentiments regarding MAPs, and to test whether – through education and awareness raising – these community sentiments can be shaped to reduce the stigma associated with MAPs, and in turn, harvest support for perpetration-prevention focused initiatives that aim to reduce the risk of CSA offending among MAPs. This led us to recent research by Jara and Jeglic (2021).

Jara and Jeglic’s (2021) study

Citing evidence suggesting that anti-stigma interventions may change attitudes toward MAPs (e.g. Jahnke et al., 2015; Kleban & Jeglic, 2012), Jara and Jeglic (2021) set out to explore public attitudes and sentiments toward MAPs, and test the effects of anti-stigma interventions on changing public attitudes toward MAPs. They employed the use of psychoeducational text, as a brief intervention, via an online survey. They also developed the Attitudes Toward Minor-Attracted Persons (ATMAP) scale to obtain posttest values and included a control group. The results were unexpected, with the survey suggesting that the public held more neutral attitudes toward MAPs than initially expected; with the intervention resulting in an increase in negative attitudes toward MAPs. Jara and Jeglic (2021) suggested that the results may have been attributed to several factors: the recruited sample (being recruited via a commercial survey company); the conditions of participation (online, in an uncontrolled environment); the use of a written text compared to an interactive intervention; or the inclusion of “myths” in the written text, which may have had a priming effect. Other limitations were noted. For example, no baseline pretesting was included and lack of demographic data meant these attitudes could not be compared by age or gender. They advocated for further research, addressing these factors, to increase the understanding of anti-stigma interventions, which may reduce the stigmatization of MAPs and enable help-seeking pathways.

On this basis, the present study sought to replicate Jara and Jeglic’s (2021) study using an Australian sample. It also sought to extend their study in three ways: (1) to include a pre-post design, (2) to examine the relative influence of gender and age on attitudes, and (3) to include a more personalized video message as an intervention. Multi-modal prevention is now advocated as best practice (Casey & Lindhorst, 2009; DeGue et al., 2014; McMahon & Seabrook, 2019). Furthermore, public health research has shown that video messaging may lead to more sustained impacts on attitudinal and behavioral change (Tuong et al., 2014) and in reducing stigma among other vulnerable populations (e.g., Barry et al., 2014), although the evidence is still mixed in terms of sexual violence prevention (DeGue et al., 2014). So, for this study, an info-video (delivered by the first author) was included and tested to see whether the mode of delivery matters was of interest. Thus, in this study we tested two hypotheses, predicting that:

(H1) Compared to the control group, reductions in participant ATMAP scores will be observed post-intervention, and across the three intervention groups

(H2) The most significant reduction in ATMAP scores will be observed in the group exposed to the video intervention.

Although age and gender were examined in the present study, this was exploratory and no predictions were made.

Method

Participants

To replicate Jara and Jeglic’s (2021) recruitment strategy, a total of 477 participants were recruited nationally via a Qualtrics XM online survey platform: 33.7% (n = 60) were located in New South Wales, 21.9% (n = 39) in Victoria and 21.3% (n = 38) in Queensland, with the remaining 23.1% in Western Australia (n = 18), South Australia (n = 13), Tasmania (n = 5), Australian Capital Territory (n = 4) and Northern Territory (n = 1). All participants were required to live in Australia and be over 18 years of age.

Half (n = 239) the participants were excluded from the study for not completing the survey or providing incomplete responses, which was a much larger proportion than that observed by Jara and Jeglic (2021; 50.1% compared to 13.6%).¹ A further 60 participants failed the manipulation check, to answer 2 of 3 reading comprehension questions correctly. As such, the final sample (n = 178) comprised of 88 males, 88 females, and 2 participants who identified as “other.” Participant ages ranged from 18 to 84 (M = 47.57 years, SD = 17.34).

Measures

This study employed the Attitudes Toward Minor Attracted Persons (ATMAP) scale developed by Jara and Jeglic (2021). The 39-item self-report survey measures an individual’s attitude toward MAPs. Responses are collected on a five-point Likert scale, ranging from (1) “strongly disagree” to (5) “strongly agree,” with higher overall scores indicating more negative attitudes toward MAPs. In the current study, a Cronbach alpha coefficient of .94 was found for the ATMAP scale, indicating a high level of internal consistency, similar to that of Jara and Jeglic (2021; α = .95). Test–retest reliability was assessed using the control group, and showed adequate reliability relative to time between testing, rho = .88 (n = 48; p < .001).

Eligibility criteria (adults [18+ years] and Australian residency) were included in the study. These were recorded dichotomously (Yes/No) to proceed. Participants were then asked to provide basic demographic data (e.g., current age, gender, location (State/Territory, cultural heritage²), before being presented with the ATMAP scale to obtain pretest scores.

Procedure

The study was implemented in line with approved ethics protocols obtained from the University of the Sunshine Coast’s Human Research Ethics Committee (A211608). Aligned with Jara and Jeglic’s (2021) methodology, participants were recruited and the ATMAP administered via Qualtrics XM, the survey development platform used by the research team, at a cost of $2,000AUD per 200 completed surveys.³ As above, participants were required to be Australian residents, over the age of 18 years, evenly distributed by gender. Similar to the sample obtained by Jara and Jeglic (2021), participants were contacted by the market research company (distributed by third-party survey platforms unidentified to the research team) they had engaged with and offered an incentive to participate in this study. The value of this incentive is unknown to the research team, and likely varied between survey platforms.

Individuals who had chosen to participate were provided with a link to the online survey platform to provide informed consent prior to commencement of the study. Participants were advised of the right to cease participation at any time, that their responses and identity would remain anonymous and confidential – with no identifiable data obtained by the research team.

Participants were asked to provide basic demographic data before being randomly allocated to one of four groups, comprising three experimental conditions and one control condition. Automatic random assignment was completed using the Qualtrics XM platform, resulting in a fairly equal distribution among groups.

Control condition. Participants in the control condition (n = 48) were immediately re-administered the ATMAP scale.

Experimental conditions. The current study adopted the minor attraction psychoeducational text developed by Jara and Jeglic (2021). This text establishes common misconceptions or myths about MAPs and presents evidence to dispel them. This evidence includes information about the nature of minor attraction, the correlation to child sexual abuse, attitudes toward treatment and treatment effectiveness – emphasizing the difference between individuals with an attraction toward minors and those who perpetrate CSA (see Jara & Jeglic, 2021 for psychoeducational text). The current study used Jara and Jeglic’s (2021) exact text for one intervention group (written myths and facts; n = 45). This text was also modified by removing the myths and retaining the facts (facts only; n = 40). The final group were presented with an “expert” video adopting the exact myths and facts script as the written condition; n = 45).

Following the intervention, participants in each of the experimental conditions were presented with the same three reading comprehension questions used by Jara and Jeglic (2021) before the ATMAP scale was re-administered presented for a second time. Using the same criteria as Jara and Jeglic (2021), participants who failed to provide two correct responses to the reading comprehension questions were excluded from analyses.

Upon completion of the study, all participants were provided with a debriefing statement and contact details for relevant support services in the event that the content had made them unsettled or distressed.

Analytic strategy

Preliminary analysis. The data were analyzed using a combination of Kolmogorov–Smirnov tests of normality, and normality and box plots to identify potential outliers. These outliers were retained after the researchers concluded they were likely naturally occurring data. ATMAP scores were assessed for normality between treatment groups, and participant gender and age. Most scores were not normally distributed. The researchers reported the conservative t-test scores with equal variances not assumed and used non-parametric tests when relevant. However, mixed between-within subject analysis of variances were conducted because of the generous sample size (>30 cases per cell; Wilson et al., 2007) and tests robustness to deviations from normality, while ensuring Levene’s Test for Equality of Variances was not violated (Pallant, 2020). Further to this, preliminary bivariate analyses (independent samples t-tests; correlations; and one-way ANOVA) were then conducted to identify whether participant demographic variables influenced ATMAP pre-intervention scale scores.

Main analysis. Mixed between-within subjects analysis of (co)variance analyses were conducted to assess the impact of four different interventions (written [myths and facts], written [facts only], info-video [myths and facts], and no intervention [control]) on participants’ scores on the ATMAP Scale across two time periods (pre- and post-intervention), controlling for the influence of age (as a continuous variable); and to assess the impact of age (coded categorically) on participants’ scores on the ATMAP Scale across two time periods (pre- and post-intervention). Post-hoc analyses were conducted to determine differences between groups.

Results

Preliminary analyses

An Independent samples t-test was used to compare scores prior to receiving an intervention between gender (excluding “other” due to insufficient numbers). There were no significant differences in pretest scores for males and females (Table 1).

Table 1. Difference in ATMAP scores by Gender.

	Male			Female			Mean difference	95% CI	df	t	p	Partial eta squared
	n	M	SD	N	M	SD
Full sample pre-score	88	133.59	22.58	88	132.48	21.26	1.11	−5.41, 7.64	174	.34	.74	<.00
Control pre	26	129.27	20.07	22	132.05	26.537	−2.78	−16.33, 10.78	46	−.41	.68	<.00
Control post	26	126.77	18.12	22	128.00	25.76	−1.23	−14.02, 11.56	46	−.19	.85	<.00
Written with myth pre	18	132.67	20.47	27	132.52	23.09	.15	−13.41, 13.70	43	.02	.98	<.00
Written with myth post	18	134.50	23.51	27	123.15	17.20	11.35	−.88, 23.59	43	1.87	.07	.08
Written no myth pre	24	137.50	23.23	15	137.53	18.70	−.03	−14.46, 14.39	37	−.01	.99	<.00
Written no myth post	24	132.21	24.53	15	133.80	16.27	−1.59	−16.12, 12.93	37	−.22	.83	<.00
Video pre	20	135.35	26.98	24	129.67	15.06	5.68	−8.18, 19.54	28.55	.84*	.41	.02
Video post	20	127.80	26.63	24	121.96	13.23	5.84	−7.58, 19.26	26.69	.89*	.38	.02
*Equal variances not assumed.

Spearman’s Rank Order Correlation was also used to explore any potential relationship between pre-intervention ATMAP scale scores and age. These analyses identified a small negative correlation between age and ATMAP scores, suggesting that older participants scored lower on the ATMAP scale than younger participants on the pre-measure (Table 2). To explore this further, a one-way between-groups analysis of variance was conducted using age as a categorical variable (coded as 1 = 18–39, 2 = 40–59, and 3 = 60+). This analysis also showed a significant difference in pre-measure ATMAP scale scores between age groups (F[2, 175] = 4.87, p = .005, η²= .05), with Tukey’s post-hoc test indicating that the mean scores for 18–39 (n = 68; M = 138.59, SD = 25.05) differed significantly to the mean scores of 60+ (n = 49; M = 126.27, SD = 18.98). Mean scores for the 40–59 (n = 61; M = 131.34, SD = 18.95) age group did not differ significantly from either of the other two groups.

Table 2. Correlation between Age on ATMAP Scores.

	Pre-Measure			Post-Measure
	n	Rho	p	n	rho	p
Full sample	178	−.218**	.003
Control	48	−.255	.081	48	−.268	.066
Written with myth	45	−.305*	.041	45	−.305	.194
Written no myth	40	−.248	.122	40	−.213	.187
Video	45	−.062	.686	45	−.152	.318

Note: **. Correlation is significant at the 0.01 level (2-tailed).

*Correlation is significant at the 0.05 level (2-tailed).

Main analyses

To test our hypotheses, a 4 (Intervention: written [myths and facts] vs. written [facts only] vs. info-video [myths and facts] vs. control) × 2 (Time: Pre- and Post- ATMAP scores) mixed between-within analysis of covariance was conducted – controlling for age as a continuous variable – to compare the impact of the four intervention conditions on participants’ pre-post ATMAP scores. Findings showed no significant interactions between type of intervention and time, Wilks’ Lambda =.99, F (3,173) = .84, p = .48, η² = .014. There was a small main effect for time alone (Wilks Lambda =.97, F [1,173] = 4.88, p = .03, η²= .03), with each group showing a reduction in ATMAP scale scores across the two time periods – including the control group (see Table 3 and Figure 1). However, the main effect comparing the three types of interventions was not significant (F [3,173] = 1.07, p = .36, η²= .02), suggesting no difference in the effectiveness of the three educational interventions. Hence, neither hypotheses 1 or 2 were supported.

Figure 1. Differences in pre and post ATMAP scores by intervention group indicating downward trends for each group.

Table 3. Attitudes toward minor attracted persons scores for treatment group and age group across two time periods.

	Treatment group	n	M	SD
Total score for pre-measure	Control	48	130.54	23.039
	Written with myth	45	132.58	21.838
	Written no myth	40	136.78	21.580
	Video	45	131.56	21.509
	Total	178	132.71	21.965
Total score for post-measure	Control	48	127.33	21.711
	Written with myth	45	127.69	20.494
	Written no myth	40	132.00	21.851
	Video	45	124.11	20.433
	Total	178	127.66	21.120
Total score for pre-measure	18–39	68	138.59	25.05
	40–59	61	131.34	18.95
	60+	49	126.27	18.98
Total score for post-measure	18–39	68	132.38	25.33
	40–59	61	127.23	16.94
	60+	49	121.63	17.92

Given preliminary analyses indicated differences in attitudes by age, a second 3 (Age category: 18–39 vs. 40–59 vs. 60+) × 2 (Time: Pre- and Post- ATMAP scores) mixed between-within analysis of variance was conducted to explore differences in the changes in ATMAP scores between age groups. However, this showed no significant interaction between age group and time, Wilks’ Lambda =.99, F (2,175) = .48, p = .62, η²= .005. There was a small main effect for time alone (Wilks Lambda =.87, F [1,175] = 27.29, p < .001, η²= .14), with each group showing a reduction in ATMAP scores across the two time periods (see Table 3 and Figure 2). Additionally, the main effect comparing the three age groups was also significant (F [2,175] = 4.74, p = .010, η²= .05). Consistent with the preliminary analyses ATMAP scores differed between age groups. Tukey’s post-hoc test indicated that the scores were significantly different between 18–39 and 60+, with higher pre- and post-measure scores in the younger group (Table 3), but that the scores for 40–59 did not differ significantly from either the 18–39 or 60+ age groups.

Figure 2. Differences in pre and post ATMAP scores by age group indicating downward trends for each group.

To replicate Jara and Jeglic, we ran analyses separately on all item measures. However, few statistical differences were identified between groups, so are not reported.

Discussion

Successful implementation of early intervention strategies requires community support. Yet, garnering support is particularly difficult where initiatives seek to address thoughts and behavior that create significant concern or abhorrence within the community, as in the case of individuals with minor-attraction toward children. Negative community sentiment creates stigma for those living with minor attraction (MAPs) by conflating minor-attraction with CSA perpetration, leading to punitive attitudes and a lack of investment in support services for MAPs; such stigma also makes it more difficult for MAPs to navigate help-seeking to manage their thoughts and reduce this potential risk. In both instances, opportunities to reduce the risk of CSA perpetration are lost yet are vital to ensure a comprehensive approach to CSA prevention. Part of the solution to these barriers is to educate the community and improve awareness of the importance, and value, of early intervention initiatives that support child safety. Importantly, to do so is not to destigmatize the act of CSA. Rather, it is to enhance community support for implementing upstream interventions to prevent CSA, alongside effective responses when it does occur, to reduce its extent and impact – consistent with a public health prevention model.

Building on recent research by Jara and Jeglic (2021), the aim of this study was to test the relative effect of different modes of educational materials on improving perceptions and attitudes toward MAPs, and the value of early intervention, as one avenue for improving community awareness and support for relevant interventions. The current study replicated, and extended, Jara and Jeglic’s study by introducing a randomized-control pre-post design, expanding the treatment conditions to include various educational messages and modalities, and assessing the relative effects of age and gender on attitudes toward MAPs.

Although the three intervention groups showed significant reductions in negative attitudes from pre to post administration, no significant differences were found across the interventions, and reductions were also observed in the control group, refuting hypothesis 1. Contrary to expectations, although the largest mean change was found in the video intervention group, this was not statistically significant compared to the written text or control conditions, refuting hypothesis 2. Interpretation of findings regarding the control group is perplexing. One explanation is that individuals opting into this study were more open to recognizing this issue, and being exposed to the study, even if briefly, may have impacted perceptions. An alternative and perhaps more likely explanation was that bias may have played a part, with participants responses to the post-measure impacted by expectations of what the study was intending to measure. That said, test–retest reliability was undertaken on the control group, and this indicated relative stability, despite the main analyses not reaching statistical significance.

The change in perceptions from pre to post intervention in all other groups indicates potential positive impacts of educational materials in changing attitudes. The lack of difference across modalities suggests that no one modality was more impactful than the other. The lack of difference in modalities was also found in Harper et al. (2018) study; perhaps then a multimodal approach is best to maximize reach and impact, rather than any one alone. Certainly, evidence suggests that multi-modal community approaches to prevention and messaging is the most influential in changing (and sustaining) knowledge and attitudes via multiple exposures to content in multiple formats (Banyard et al., 2018, 2022; Casey & Lindhorst, 2009; DeGue et al., 2014; McMahon & Seabrook, 2019). In any case, the present findings are inconsistent with Jara and Jeglic’s (2021) findings, and – given similar sampling methods were used – goes some way to alleviate concerns raised by them regarding potential negative consequences of educating the public and conflating MAPs with pedophilia and/or child sexual offending, inadvertently reinforcing myths and inducing negative attitudes and labeling (see Jahnke et al., 2014).

It is possible that differences between the present findings and Jara and Jeglic’s (2021) research reflect potential differences in overarching philosophies toward justice and rehabilitation, which may more broadly influence public perceptions in each country. However, recent comparative research indicates similar, punitive attitudes toward crimes such as rape and child sexual abuse among community members in Australia and the US (De Soto et al., 2022). That said, similar to Jara and Jeglic (2021) Harper et al. (2022) found purely fact-based education increased immediate negative effects in terms of increased perceptions of deviance, leading them to conclude that intervention relying only on factual information may inadvertently reinforce the negative perceptions of minor attraction. Our present findings are more consistent with previous literature examining attitudinal change (Holden et al., 2014; Jahnke et al., 2015; Kleban & Jeglic, 2012), reinforcing the potential value of community awareness and education initiatives to enhance public knowledge and understanding of sensitive community issues, but these differences do raise questions regarding the design and messaging associated with anti-stigma interventions. As Lawrence and Willis (2021) note, messaging must address the “nature of choice and controllability of other sexual preferences” (p. 157), to deconstruct the presumption that minor-attraction inevitably leads to harmful sexual behavior, and reinforce that behavior can be managed with support.

The pattern between participant age and attitudes toward MAPs was an unexpected finding. Given the landscape for open discussion and tolerance about sex, sexuality, and inclusion among millennials (e.g., Janmaat & Keating, 2019), this seems a somewhat counterintuitive finding. What this might suggest, however, is that different messaging platforms (and targeted messaging) for different age cohorts within the community may be required to maximize impact of anti-stigma and prevention messaging, reinforcing a multi-modal approach to prevention practice. The fact that no difference was found among gender means that tailored messaging across gender may not be as important as the consideration of age.

As Harper et al. (2022; see also Paluck et al., 2021) outline, there are various “clusters” of anti-stigma interventions. The anti-stigma intervention used in the present study is psychoeducational in nature, focusing on the “cognitive and emotional understanding of out-groups” (p. 947), by challenging myths and stereotypes and balancing this with factual information to help enhance understanding of minor-attraction. Although anti-stigma reduction studies focused on complex social problems indicate that using vignettes and education sessions can produce measurable effects in the immediate time period following exposure (e.g., Morgan et al., 2018, 2021), the longevity of these brief interventions in influencing participants’ attitudes, and hence sustained benefits, is relatively unclear (Paluck et al., 2021). Regarding minor-attraction, recently Harper et al. (2018) introduced the process of narrative humanization, that is, “presenting information about stigmatized people in the form of a first-person narrative” (p. 535) as one way to positively influence perceptions and attitudes in anti-stigma interventions. This form of intervention combines psychoeducational components with first-person exposure (albeit indirectly) through personalized stories to break down stereotypes and barriers. Underlying these ideas is the premise that attitudinal change requires more than a shift in cognitions and thinking brought about by education and awareness, to challenge the underlying negative affective responses to minor-attraction (and MAPs), via humanizing techniques (Harper et al., 2019; Lawrence & Willis, 2021). In their study, it was hypothesized that a combination of psychoeducation and first-person narratives might create more meaningful and longer-term change in attitudes and beliefs (Harper et al., 2022). Building on earlier work (Harper et al., 2018) and using a community sample, Harper et al. (2022) compared two anti-stigma interventions, both 5-minute videos, one scientific fact-based video delivered by an expert, and the other using narrative humanization techniques. Although some variations were noted, overall, they found that both types of presentations produced immediate reductions on most measure of stigma; and while rebound effects were evident in the four-month follow-up period the effects were still significant. Most notably, although reduced stigma was more pronounced in the information video in the immediate period following exposure, the narrative condition did not produce the same level of bounce-back, meaning that the longevity in attitudinal change may hold longer in narrative condition.

A recent systematic review undertaken by Lawrence and Willis (2021) showed that informative interventions aimed to educate and enhance empathy can indeed be effective in challenging mainstream perceptions, feelings (e.g., anger) and reducing punitive attitudes toward punishment. Overall, the systematic review highlighted the need for humanization narratives as part of anti-stigma interventions to maximize impact, challenge implicit assumptions, and create sustained attitudinal change which “may lead to more accepting societal attitudes” (Lawrence & Willis, 2021, p. 155).

Limitations and future directions

There are some limitations that warrant discussion, but also offer future research opportunities. First, although randomized assignment was used, to replicate Jara and Jeglic’s study, this research was conducted online, limiting the capacity to control for other extraneous variables that may have impacted participants’ concentration and focus. Although parameters were put in place to enhance reliability of respondents (manipulation tests, minimum time parameters), this may not have been sufficient to counteract distractibility or inattention (see also Jara & Jeglic, 2021). Second, by using a similar recruitment strategy as Jara and Jeglic, the representation of participants’ views sourced via a marketing platform may not easily generalize to the community population. Further, it was a paid online sample, so testing this in an unpaid community sample will help discern the overall impact of educational interventions to improve community awareness and perceptions about MAPs more generally. Jara and Jeglic also raised the need for surveying a broader community population. However, given the sensitivity of this issue, surveying the public online also comes with increased risks requiring significant planning to implement safely, and needs careful consideration.

Third, a priori power analysis (conducted using G*Power v3.1.9.2; Faul et al., 2007) indicated that a minimum of 178 and 204 participants respectively were required to achieve 80% power to detect small to medium effects (f = 0.25) for the analysis of variance tests. Despite the overall sample size and cell sizes being adequate for conducting these tests (Wilson et al., 2007), attrition meant that intervention group sizes were smaller than anticipated. Given the nonsignificant findings in this study, testing these effects again with a larger sample size may reduce the potential for type 2 error.

Fourth, findings from this study demonstrate an immediate effect on perceptions and attitudes only. There was no capacity to follow up participants to test the sustainable impacts of the educational messaging over time. There is need for more research to examine longer term effects of educational messaging and its influence on attitudes, including whether these interventions result in increased support for prevention and early intervention initiatives (Harper et al., 2022). It is likely that multi-modal, and more intensive educational campaigns will be required to have any meaningful effect (DeGue et al., 2014). Indeed, Paluck et al. (2021) note that “light-touch interventions” that are brief, easy and cheap to implement, currently dominate prejudice reduction strategies, but the longer-term effects of these interventions are yet to be established. Certainly, repeated interventions that challenge individuals at both cognitive and affective level are likely required to produce more long-lasting attitudinal change (Lawrence & Willis, 2021); and, that “we might expect repeated exposures to such humanizing messaging to have a more profound and lasting effects in a way that informative presentations may not” (Harper et al., 2022, p. 956).

Finally, the educational text provided was gender-neutral. It would be interesting to test for differences in participant responses, according to whether the person with minor-attraction toward children is male, female or nonbinary. Moreover, it needs to be acknowledged that cultural differences may exist between countries, and that a larger study that includes a cross-cultural comparison may help to determine more nuanced and locally tailored early intervention initiatives. These are all fruitful avenues for future research to help continue to the momentum for, and social acceptance of, upstream initiatives as part of a comprehensive public health approach to combatting CSA.

Conclusion

Current public and professional misconceptions of MAPs have likely contributed to the lack of uptake and support for perpetration-focused early intervention initiatives to date. As Assini-Meytin et al. (2020) point out, community sentiments and public reactions influence policy professionals and decision-makers, oftentimes leading to a focus on punishment and deterrence after the fact, rather than investment in proactive strategies that seek to prevent CSA from occurring in the first place. This is both short-sighted and potentially harmful by omitting opportunities to intervene early, given CSA is a preventable social problem. The findings in this study raise hope that community education and messaging initiatives may help change misconceptions and attitudes and improve public acceptance of the value of early intervention, to support the introduction and successful rollout of early intervention services in new locations. Certainly, educating the public, key policy professionals and decision-makers who wield the power to support and implement early intervention measures may prove valuable in these endeavors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors reported there is no funding associated with the work featured in this article.

Notes on contributors

Nadine McKillop

Nadine McKillop, Associate Dean (Research), Senior Lecturer in Criminology and Justice, and Co-Leader of the Sexual Violence Research and Prevention, Unit, School of Law and Society, University of the Sunshine Coast, Sippy Downs, Australia.

Stephanie Price

Stephanie Price Candidate and research assistant, Sexual Violence Research and Prevention, Unit, School of Law and Society, University of the Sunshine Coast, Sippy Downs, Australia.

Notes

¹ The poor attrition in the Australian sample, and compared to the US sample, was unexplained. However, the Qualtrics (2023) suggests that sensitive content may reduce survey completion and could account for the high attrition rate found in our study.

² Only 1 participant identified as Aboriginal and Torres Strait Islander, preventing the use of cultural heritage as a variable in any analyses.

³ Incomplete surveys were provided to the research team at no extra cost.