Abstract
Child emotional abuse is repeated parental behaviour that conveys to the child they are worthless, unloved, unwanted or only of value in meeting another’s needs. It is widespread and associated with mental health harms — including depression, anxiety, self-harm, and suicide attempts — at levels comparable to those linked to physical and sexual abuse. However, there appears to be a lack of understanding of its impact and the urgency in preventing its occurrence. This systematic review aimed to synthesise findings from peer-reviewed studies published since 2000 that examined public knowledge and attitudes about the emotional abuse of children aged 0–18 perpetrated by parents. Of 39 included studies, only 4 were rated high-quality. Most often, studies examined recognition of child emotional abuse or its acceptability. Few studies investigated willingness to intervene or knowledge about prevalence or its harm to psychological health. Recognition of child emotional abuse varied depending on the type of behaviour and was highest for repeated hostility behaviours. Generally, child emotional abuse was less recognised and perceived as less harmful than sexual or physical abuse. There is a need for more high-quality data on public understanding of child emotional abuse to inform educational strategies for improving its prevention and response.
Background
The United Nations Convention on the Rights of the Child states that children have the right to be protected from all forms of violence, including mental violence (United Nations General Assembly, 1989). There has been relatively less attention invested in the prevention of emotional abuse in contrast to physical and sexual abuse (Brassard et al., 2020). Child emotional abuse has been defined as parental behaviour, typically repeated, that conveys to the child they are worthless, unloved, unwanted or only of value in meeting another’s needs (Mathews et al., 2021). It is also referred to as emotional maltreatment or psychological abuse. It is exemplified by acts of hostility, terrorising, rejection, isolation, corruption and denying emotional responsiveness. While historically emotional neglect was often considered a form of child neglect, there is a growing consensus that it is better conceptualised as a form of emotional abuse (Mathews & Dube, 2025). Emotional neglect is defined as emotional unresponsiveness in child interactions and isolation (caregiver acts that consistently and unreasonably deny the child opportunities for interacting and communicating with peers or adults inside or outside the home) (Brassard et al., 2020). Child emotional abuse is a major public health problem. Prevalence estimates vary but are universally high across continents. Stoltenborgh et al. (2015) found a pooled estimate of 36.3% (95% CI 28.1% - 45.4%) for self-reported abuse across 42 studies, not including emotional neglect. This was higher than the prevalence of sexual abuse (12.7%) and physical abuse (22.6%). These high estimates show no signs of decreasing, despite recent evidence that physical abuse and some forms of sexual abuse show signs of decline (Mathews et al., 2023). In addition to its high prevalence, it has a substantial negative impact on psychological health (Norman et al., 2012). For example, a recent nationally representative survey of 8,503 Australians found that adults who experienced emotional abuse in childhood were twice as likely to report mental disorders (depression, generalised anxiety disorder, or PTSD) self-harm, or attempted suicide (Lawrence et al., 2023; Scott et al., 2023). These risks were as high as those from child sexual abuse and remained after controlling for demographics and other co-occurring forms of child abuse, which themselves confer substantial risk.
Despite clear research evidence of the high prevalence and harmful impact of emotional abuse, its prevention and intervention remains challenging. Professionals who are mandated to report suspected cases of child abuse report less confidence in identifying and responding to emotional abuse, given its less overt nature and absence of physical evidence (McTavish et al., 2017). In comparison with other forms of child abuse, there is less of a clear demarcation between sub-optimal parenting behaviours and emotionally abusive behaviours. Even child protection professionals show low levels of recognition of many forms of emotional abuse and appreciation of its harm (Baker et al., 2021). Furthermore, professionals perceive that some emotionally abusive behaviours are more acceptable in the community than physical or sexual abuse, and can be reluctant to intervene (Baker et al., 2024). Baker et al. (2021) argue that clinicians and the public may be unaware of the strong evidence that exists on the harm from emotional abuse and that there is a need to change social norms around harsh and insensitive parenting.
Changing social norms requires effort at a societal and community level in addition to focusing on professionals who work with children and families. Given its high prevalence, it is likely that many adults know a child who is experiencing emotional abuse. Understanding the knowledge that community members hold about child emotional abuse is important to inform approaches to its prevention. For instance, education programmes for children and parents are an important strategy in the prevention of child sexual abuse (Russell et al., 2024), as is changing norms about corporal punishment for the prevention of physical abuse (Havighurst et al., 2023). The field of mental health has built on research related to mental health literacy (defined as knowledge and beliefs about mental disorders which aid their recognition, management or prevention; Jorm et al., 1997) to improve recognition of different mental health problems, knowledge about effective treatments, and how to seek professional help or give support (Jorm, 2012). Similarly, for the prevention of emotional abuse, it is important to research how well the public understands the problem of emotional abuse, can recognise when it is occurring, and are willing to respond or intervene. This research can then be used to inform interventions to target incorrect knowledge or beliefs and change social norms about the acceptability of emotionally abusive parenting. While the extant literature includes systematic reviews on public knowledge and attitudes towards child sexual abuse (Babatsikos, 2010; Buller et al., 2020; Sawrikar & Katz, 2017) and child abuse in Saudia Arabia specifically (Sheikh et al., 2022), there is no synthesis of studies specific to child emotional abuse.
The aim of this review therefore was to synthesise evidence on public knowledge and attitudes regarding child emotional abuse. Specifically, we sought to explore how the public conceptualises emotional abuse and identifies its occurrence, their awareness of its prevalence and psychological impact, and their beliefs about its acceptability, causes, and willingness to intervene.
Methods
The systematic review protocol was registered in Prospero (CRD42024495431).
Eligibility Criteria
Studies were eligible for inclusion if they investigated knowledge and attitudes of the public towards child emotional abuse perpetrated by parents and caregivers. The following inclusion criteria were used: (1) Studies required an assessment of child emotional abuse knowledge or attitudes, including knowledge related to child emotional abuse; beliefs about factors influencing emotionally abusive behaviour, societal norms; and attitudes such as willingness to act or facilitators or barriers to responding to child emotional abuse. Studies could focus on child emotional abuse in isolation or compare child emotional abuse to other forms of child abuse. (2) Participants were members of the public, including parents and carers as well as children and young people. (3) Studies were published in English after the year 2000 (4) Studies were peer-reviewed and used quantitative methods (such as surveys and questionnaires), including intervention studies with baseline data.
We excluded studies based on the following criteria: (1) Child emotional abuse was not clearly distinguished from other forms of child abuse (2) The perpetrators of child emotional abuse were peers or other adults in the community. (3) Participants were registered health professionals, teachers and police officers (who are often mandatory reporters) (4) Studies used qualitative methods or were reviews, books, dissertations or commentaries
Information Sources
Three online databases (PsycINFO (Ovid), MEDLINE (Ovid), ERIC (EBSCO)) were searched on 26 November 2024 for studies published in English between 2000 and 25 November 2024. References of the included studies were also checked for additional papers not found in the initial search.
Search Strategy
An initial scoping search of PsycINFO and Google Scholar was undertaken to identify key articles on the topic and to assist in informing keywords. A comprehensive search strategy was developed for each database using both controlled vocabulary and text words. The final search strategy for each database can be found in Appendix 1.
Study Selection
Study citations from all searches were imported into Covidence (Version 2). After duplicates were removed, the first 100 articles were screened by two reviewers. As there was only fair to moderate agreement (kappa = 0.39), discrepancies were discussed by the research team until consensus was reached, and a second set of 40 articles were independently screened (kappa = 0.75), indicating substantial agreement. The remaining titles and abstracts were screened independently by a single reviewer (ET or CT). The full text of potentially relevant citations was then assessed in detail against the inclusion criteria by two independent reviewers (ET, AM, CT, QC). CT was the primary reviewer, with the second review split amongst the remaining three reviewers for consistency. Conflicts in screening were resolved by either discussion between the pairs or with the research team until a consensus was reached.
Data Extraction
A data extraction tool was developed and piloted among all reviewers. To ensure internal consistency, reviewers tested a sample of 3 articles in pairs and compared their results before finalising the data extraction form. Data were extracted on the characteristics of each included article (title, author, date of publication, aim/objective, study design/method), sample (type of population, sample size, demographics, country, income level according to the World Bank), the assessment tool, and a summary of key findings for knowledge, beliefs, and attitudes. When studies reported data separately for child emotional abuse and emotional neglect, both were extracted as forms of emotional abuse. Conflicts between reviewers were resolved in pairs, or with the research team where consensus couldn’t be reached between the pairs.
Quality Assessment
The quality of studies was assessed using the Joanna Briggs Institute critical appraisal checklist for studies reporting prevalence data (Munn et al., 2014). Each paper was assessed independently by two review authors and discrepancies were resolved by consensus between pairs or by a third author (AM) if necessary. An overall study quality score was derived based on scores of 0–3 (low), 4–6 (medium) and 7–9 (high).
Data Synthesis
Data was synthesised narratively. The common themes or findings across studies were analysed, and the results are described below in relation to the review aims.
Results
Study Characteristics
A total of 5,825 articles were identified as potentially relevant after duplicates were removed. After title and abstract screening, 4,749 studies were excluded based on the child abuse type or because they were focused on prevalence or outcome studies of child abuse. In full-text screening, 100 studies were excluded, leaving 39 studies meeting all inclusion criteria (see Figure 1). PRISMA Flow Diagram for Study Selection.
Summary of Study Characteristics
Studies met between 0 and 9 of the quality appraisal items (M = 4.33, SD = 1.85), see Figure 2. Four studies were rated high quality, with only one study meeting all criteria in the quality assessment (Bensley, Ruggles, et al., 2004). Where studies lacked in quality included approaches to the sampling frame and recruitment strategies which often were not reported in sufficient detail. Only 9 studies used probabilistic sampling methods. Instead, convenience sampling methods were commonly used, and accordingly, the response rate was often not reported (66.7%) nor how well the sample represented the target population. It was also unclear if the sample size was sufficient for most studies, as a sample size calculation was not reported (79.5%). Less than half of the studies used appropriate statistical analysis, as often key information such as confidence intervals, standard deviations, or standard errors were not reported. The strengths of studies in this review included study participants being described in adequate detail, and measuring child emotional abuse in a consistent way with measures that were often piloted or tested prior to conducting the study. Summary of Quality Assessment - The Joanna Briggs Institute Checklist for Prevalence Studies
Participant Characteristics
Of the 39 studies, 3 included children or adolescents only, 2 included both children and parents/carers, 16 included parents/carers only, 17 sampled the general public, and 1 sampled pastors. Three studies did not report the gender of participants. For the remaining, the percentage of male participants ranged from 0% to 86.2%, median = 43.7% (IQR 22.4). Of the 36 studies that included adult participants, 18 studies (50.0%) used a parent-only sample, 8 studies reported that parents comprised 10.2% to 78.9% of the sample, and 10 (27.8%) did not report how many were parents. Most studies (33/39, 84.6%) did not report whether participants had experienced a history of child abuse. For the 6 studies where this was reported, estimates ranged from 13.8% to 85.9%.
Measurement
The focus of most studies was on multiple forms of child abuse, including emotional abuse. Only two studies focused on child emotional abuse specifically (Gulirmak & Orak, 2021; Uslu et al., 2010). The measurement of knowledge and attitudes related to child emotional abuse was measured in a variety of ways. These varied in whether the label of emotional abuse was used or whether brief vignettes or descriptions of parenting behaviours were used. More than half of the studies (21/39, 53.8%) developed their own measure, 15 reported adapting an existing tool (38.5%), and only 6 (15.4%) used an existing tool (some studies used multiple tools). The most commonly used tool was the Perception of Child Maltreatment Scale (Fakunmoju & Bammeke, 2013), used in four studies. Three studies used the child abuse vignettes from Giovannoni and Becerra (1979), two studies used the Recognition of Emotional Maltreatment Scale (Uslu et al., 2010), and two studies used the Child Neglect and Abuse Awareness Scale for Parents (Ünal & Boz, 2020).
Most commonly, studies investigated knowledge about what constitutes emotional abuse in children (k studies = 24) or the acceptability of emotionally abusive behaviours (k = 12). There were fewer studies investigating knowledge of impact on psychological health (k = 6), willingness to act (k = 6), knowledge of prevalence (k = 2), attitudes related to taking action (k = 2), and none that investigated beliefs about causal influences on child emotional abuse. Full details of each study’s outcomes are found in Appendix 3.
Findings
Recognition of Child Emotional Abuse
Twenty-four studies investigated knowledge or beliefs about what behaviours constitute child emotional abuse. Most commonly (14/24, 58%), studies presented a variety of emotionally abusive behaviours and assessed participants’ beliefs on whether each constituted maltreatment or child abuse. Only one study asked whether behaviours were specifically “emotional abuse” (Fakunmoju et al., 2013). Beliefs about whether behaviours constituted child abuse varied across studies. As behaviours were not described consistently across studies, they were grouped into types of emotional abuse: hostility, terrorising, rejection, isolation, corruption, and denying emotional responsiveness.
Hostility behaviours, such as “Repeatedly calling a child “stupid” or other names”, “Verbally abusing, cursing, or calling a child horrible names”, “A parent who regularly calls their child names and embarrasses the child in front of others”, frequently had high recognition rates (>90%) as abuse (Al-Hassan & Lansford, 2011; Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004; Bright et al., 2022; Price et al., 2001). Recognition was lower for “A parent constantly yells at a child” (72%) (Tucci & Mitchell, 2022), and descriptions of behaviours that did not include a repeated element, such as “Public shaming” (62%) (Alsalman et al., 2023) and “Calling child useless” 30% (Lui et al., 2019).
Recognition of terrorising behaviours as abuse tended to be greater when the behaviour included indicators of greater severity. For example, 92% endorsed “Threatening to seriously hurt the child” (Bensley, Ruggles, et al., 2004) but only 48% endorsed “Threatening with physical violence” (Alsalman et al., 2023). Endorsement of threatening to abandon a child was also above 90% in two studies (Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004) but only 49% in Lui et al. (2019).
Across studies, there was variation in the endorsement of rejection behaviours as abuse, from 64% for “Telling child you don’t love them”(Alsalman et al., 2023), 80% for “Telling a child that he or she is responsible for making the parent suffer in important ways” (Bensley, Ruggles, et al., 2004), to above 90% endorsement for “Treating a child in a way that makes him/her feel unwanted” (Bammeke & Fakunmoju, 2016) and “Blaming a child for parent’s problems” (Fakunmoju et al., 2013).
Isolation behaviours, such as “Keeping a child isolated from contact with anyone but the parent” and “Locking a child younger than 11-year old alone in a room a whole day as a form of discipline” were generally endorsed by 80%–90% of participants across included studies (Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004; Fakunmoju et al., 2013; Son et al., 2017; Tucci & Mitchell, 2022), with the exception of Lui et al. (2019), where only 64% endorsed “Locking child outside the house” as constituting child abuse.
Recognition of corruption behaviours, such as “Engaging in ‘dirty’ discussions and sexually explicit conversations to a child’s discomfort”, was in the range of 80%–90% or more across included studies (Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004; Fakunmoju et al., 2013; Mesman et al., 2020; Price et al., 2001; Rhee et al., 2003).
Recognition of denying emotional responsiveness as child abuse, such as “Withholding love and affection from a child” and “Parents completely ignore their child and don’t care what he does” was also generally in the range of 80%–90% or more, with the exception of “Never hugging child” which was only endorsed as abuse by 12% (Lui et al., 2019).
Correlates of Recognition
Eleven studies investigated whether recognition of child emotional abuse varied according to sociodemographic characteristics or the experience of emotional abuse as a child. Four studies investigated whether recognition differed according to gender, with three medium-quality studies finding that women had higher recognition than men (Price et al., 2001; Uslu et al., 2010; Ünal & Boz, 2020) and one low-quality study not finding a difference (Korbin et al., 2000). While there was no difference according to socioeconomic status for American adults (Price et al., 2001), two studies with Turkish parents found that recognition was higher amongst higher income families, but the effects of education and parent age were inconsistent (Uslu et al., 2010; Ünal & Boz, 2020). One low-quality study found that Israeli children had greater levels of recognition than parents (Lev-Wiesel et al., 2020). The effects of ethnicity or country of residence on recognition of emotional abuse was inconsistent. A low quality study of adults from America, Ghana, and Nigeria found White/Caucasian respondents had higher recognition than Black/non-Caucasian respondents for some abusive behaviours (Fakunmoju et al., 2013), whereas recognition was similar for American, Korean, and Japanese mothers (Son et al., 2017) and ethnicity was not a factor for American adults (Korbin et al., 2000).
Three studies largely found that experiencing child emotional abuse was not related to recognising it as maltreatment (Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004; Moon et al., 2019).
Comparison to Other Forms of Child Abuse
Eight studies compared recognition of emotional abuse to other forms of child abuse. In almost all studies, descriptions of sexually abusive behaviours had the highest rates of recognition as constituting child maltreatment or abuse (Bammeke & Fakunmoju, 2016; Bensley, Ruggles, et al., 2004; Chan et al., 2008; Lee et al., 2014; Price et al., 2001). For example, a high-quality study of American adults found that five of the six behaviours with the highest rates of recognition were for sexual abuse (Bensley, Ruggles, et al., 2004). The only exceptions were Bammeke and Fakunmoju (2016), which found universally high recognition of all types of child abuse, Korbin et al. (2000), which found that physical abuse was most frequently mentioned, and Hoefnagels et al. (2021), which didn’t ask about sexual abuse as the study was with children. Recognition of emotional abuse was similar to physical abuse but higher than neglect.
Knowledge of Prevalence
Two low-quality studies, one in Australia (Tucci & Mitchell, 2022) and one in Saudi Arabia (Alsalman et al., 2023), reported on the perceived prevalence of different forms of child abuse, with contrasting findings. The Australian sample believed that sexual abuse and physical abuse occurred the most frequently with emotional abuse and neglect being least prevalent (data not reported). In contrast, it was more frequent for emotional abuse (32%) and neglect (41%) to be perceived as the most common form of abuse in the Saudi Arabian study, compared with sexual abuse (5%) and physical abuse (22%).
Knowledge of Impact on Psychological Health
Three medium-quality studies reported that a majority of participants agreed that emotional abuse can impact psychological health. Seventy-six percent of a Canadian sample believed that children can often develop psychological problems from emotional abuse (Clement & Chamberland, 2007), which was similar to the 71% of Qatari parents who agreed that “yelling affects child growth and development” (Hendaus et al., 2020). There was also high recognition in a study of Indian adults, which reported that 95% of the sample agreed that children affected by emotional abuse should get counselling (Priyadarshini et al., 2023). Despite this relatively widespread recognition of the impact of emotional abuse, three studies reported that it was perceived as less harmful than physical abuse. Hendaus et al. (2020) report that two-thirds of participants agreed that yelling is less harmful than hitting. Mesman et al. (2020) and Woudstra et al. (2021), which overlapped in their multi-country samples, reported that emotional abuse was perceived as less harmful to children than physical abuse, and that emotional neglect was less harmful than emotional abuse, with the size of the difference varying between countries. There were also cultural differences in Son et al. (2017), which found that compared to American mothers, Korean mothers believed emotionally abusive behaviours would be more likely to cause immediate psychological harm and longer-term psychological harm.
Beliefs About Social Norms
Twelve studies measured participants’ beliefs about the acceptability or perceived severity of behaviours describing child emotional abuse. Three studies assessed a variety of emotionally abusive behaviours (Choi & Thomas, 2015; Lui et al., 2019; Pierce & Bozalek, 2004). Perceived acceptability varied according to the specific behaviour, with two studies reporting the most serious form to be “The parents are constantly screaming at their child, calling him foul names” (Choi & Thomas, 2015; Pierce & Bozalek, 2004). Beliefs about acceptability appeared to be lower in the study of South African adults (Pierce & Bozalek, 2004) than in the study with Korean parents (Choi & Thomas, 2015). Beliefs about acceptability also varied by country in Wadji et al. (2023), which reported lower acceptability in Canadian adults compared with adults from Cameroon. This study also reported higher acceptability of emotional abuse than sexual abuse, physical abuse and neglect. Significant minorities of parents in Hendaus et al. (2020) and Sharma and Mathur (2019) reported beliefs in the acceptability of some forms of child emotional abuse: 30% agreed that parents have the right to yell or hit their children (Hendaus et al., 2020) and 20% agreed it was okay to verbally offend, yell, humiliate or create the feeling of unworthiness in the child as form of punishment (Sharma & Mathur, 2019). While there were high ratings of unacceptability found in a sample of Norwegian adults for a family blaming the child for parent conflict, beliefs about unacceptability were higher in females and those with lower education (Lovlie, 2023). Females also had higher beliefs about unacceptability in a study of American parents (Ferrari, 2002). In contrast, acceptability did not vary by gender, age, marital status, education or number of children in Wadji et al. (2023).
Two low-quality studies sought to understand the views of children on child emotional abuse (Choi & Thomas, 2015; Massarweh & Kosher, 2023). Similarly to adults, the perception of seriousness or severity varied depending on the specific type of emotional abuse, but both showed evidence that girls thought emotional abuse was more serious or severe than boys. In addition, Korean children perceived emotional neglect to be less serious than other forms of neglect (Choi & Thomas, 2015), and Jewish children rated emotional abuse as more severe than Arab children in a study in Israel (Massarweh & Kosher, 2023).
Two studies assessed beliefs about the acceptability of the respondent’s own reported history of experiencing emotional abuse as a child (Lee & Kim, 2011; Ni & Hesketh, 2021). Both studies recruited large samples of Chinese or South Korean young adults who completed the ISPCAN. A significant proportion of respondents believed that the emotional abuse was reasonable and justified (41.4%; Ni and Hesketh (2021)) or reasonable and justified or never on purpose (34.3%; Lee and Kim (2011)). Furthermore, many respondents (61%) in Ni and Hesketh (2021) believed that they had experienced less emotional abuse than their peers.
Willingness to Take Action
Three medium-quality studies examined the willingness to take action in response to various descriptions of emotionally abusive behaviours (Benbenishty & Schmid, 2013; Chan et al., 2008; Rhee et al., 2003). These found willingness varied depending on the behaviour described, even within types of emotional abuse. A study of Korean American pastors compared different forms of corruption and reported that 15.3% of respondents would report a family to child protective services if the parents encouraged their child to steal from a market, but 50% would report where parents use drugs in their 8-year old’s presence (Rhee et al., 2003). Benbenishty and Schmid (2013) found relatively high willingness to report if a mother keeps calling her 5-year-old foul names, but equivocal willingness where a father repeatedly warns his 8-year-old that he will kill him with beatings if he does not obey his mother. These studies also found that willingness to act was correlated with the belief that the behaviour constituted abuse, but there remained substantial variance unaccounted for (correlations varied from 0.43 to 0.67). A low-quality study of Norwegian adults investigated likely response to a case of emotional abuse, and reported about a third would do nothing or wait until the situation worsens, about a fifth would act if the parents agreed and nearly half would act even if the parents disagree (Lovlie, 2023). The only sociodemographic characteristic associated with increased severity of response was female gender.
Of the three studies that compared willingness to report emotional abuse with other forms of child abuse, all found that emotional abuse was the least likely to be reported (Benbenishty & Schmid, 2013; Bensley, Simmons, et al., 2004; LeCroy & Milligan-LeCroy, 2020). A high-quality survey of American adults found that almost all would be likely to report sexual abuse (95%), and a majority would report physical abuse (84%) and neglect (68%), but only 40% would report emotional abuse (Bensley, Simmons, et al., 2004).
Attitudes About Taking Action
Two medium-quality studies reported on attitudes that could help or hinder acting in response to emotional abuse (Hoefnagels et al., 2021; Rhee et al., 2003). A study of 864 Dutch children assessed attitudes to talking about different forms of child abuse described in vignettes and found no difference in the “talkability” of emotional abuse compared to physical abuse or neglect (Hoefnagels et al., 2021). A study of 87 Korean American pastors reported that 21% believed emotional abuse cases are easy to prove versus 79% who disagreed (Rhee et al., 2003).
Discussion
This systematic review aimed to systematically synthesise evidence on public knowledge and attitudes regarding child emotional abuse. Although we identified a reasonable number of studies, most of these focused on the recognition of various emotionally abusive behaviours as child abuse. There is far less research that has examined beliefs and attitudes about intervening, and none that has examined beliefs about causes of emotional abuse, including who is most at risk. These outcomes were more often investigated for child abuse as a whole and could not be extracted for emotional abuse specifically. Furthermore, few studies were rated high in quality, as many recruited convenience samples with questionable representativeness. Nevertheless, consistently across studies, there were significant minorities of respondents who did not recognise emotionally abusive behaviours as child abuse or believed that they were acceptable. Overall, review findings suggest that there is room for improvement in public knowledge, beliefs, and social norms related to child emotional abuse.
A challenge in synthesising the findings across studies was the lack of consistency in measurement tools, compounded by the various forms of emotional abuse and different definitions that include or exclude emotional neglect. Definitional issues are not new (Slep et al., 2022), and this review showed that knowledge and beliefs vary depending on the description of the emotional abuse. For example, recognition of behaviours as abuse varied between 12% and 64% in Lui et al. (2019) for never hugging a child versus locking a child outside the house. This is important to note as it raises the question of how to understand findings from studies that only use a label of emotional abuse, which is open to interpretation by respondents. Given this, future research should be clear on how emotional abuse is defined so that findings can be compared across studies. Furthermore, the development of a standard set of vignettes that describe common presentations of child emotional abuse may be helpful to move the field forward.
The included studies were diverse in terms of the countries where they were conducted and the target populations they focused on. However, no studies were identified from low-income countries, and studies involving children or adolescents were less common. Analysis of whether knowledge and attitudes vary according to sociodemographic factors was largely inconsistent. While some studies observed between-country differences, these were not always consistent and varied depending on the measure. For example, Son et al. (2017) found Korean mothers were more likely to endorse beliefs about psychological harm from emotionally abusive behaviours, but American mothers believed they were less acceptable. However, there was some evidence that females had better knowledge and attitudes than males, regarding recognition, beliefs about acceptability, and willingness to intervene. The effect of other correlates, such as level of education or income, was inconsistent. Further work is needed to understand whether there are population sub-groups that should be a priority for education about child emotional abuse. There was also evidence that willingness to act was influenced by recognising the behaviour was abuse. This is consistent with other reviews of influences on mandatory reporting (Wilson & Lee, 2021), which found that lack of knowledge of child abuse was an important influence alongside systemic factors, sociocultural context, and reporter traits. Included studies also largely ignored responses other than alerting authorities, such as approaching the family to raise the issue or to recommend interventions that provide parenting support. The influences on these alternative intervention options remain unexplored.
There were some clear differences in knowledge and beliefs/attitudes about emotional abuse compared with other forms of child abuse. Emotional abuse was less well-recognised than sexual abuse, was perceived as less harmful than physical abuse, and was less likely to be reported than other forms of child abuse. Together, these findings provide support for the perception that has been noted by others that emotional abuse is perceived as more acceptable than sexual or physical abuse (Baker et al., 2024). As research has shown that emotional abuse poses a risk to mental health that is comparable to, or even greater than, other forms of child abuse, this public misconception warrants attention (Gardner et al., 2019; Norman et al., 2012; Scott et al., 2023).
A notable gap was research on knowledge about prevalence. Data show very high levels of emotional abuse in the community (Mathews et al., 2023; Stoltenborgh et al., 2015), but it is unclear how much the community is aware of just how common it is. There are parallels to the mental health field here. Historically, mental health problems (such as depression) were not understood to be common, and certainly not thought of as very impactful on public health. The Burden of Disease studies beginning in the 1990s first illuminated the high prevalence and burden of mental health problems, in particular depression (Murray et al., 1996). Consequently, much work has been done over the past two decades to inform the community about its prevalence and increase awareness of signs and symptoms and how to seek help or support others. This has been associated with substantial increases in depression literacy and uptake of professional treatments (Jorm et al., 2017). Similarly, with accumulating evidence that the prevalence of emotional abuse is high and a major contributor to poor mental health, is there a need for community campaigns to increase recognition, appreciation of its harm, and intentions to prevent or respond to it? This could complement the call for strategies to improve the uptake of parenting programmes that teach evidence-based parenting and emotion regulation strategies, including campaigns to reduce stigma and normalise participation (Doyle et al., 2023; Sanders et al., 2018). These parenting programmes teach alternatives to harsh and insensitive discipline, and some have shown they lead to reductions in child maltreatment substantiations (Sanders et al., 2024). Community campaigns should be rigorously developed and tested before implementation to optimise their contribution to preventing and responding to child emotional abuse and to avoid unintended harm.
This review has some limitations to note. We only included studies published in English, and it is possible that we may have missed some additional studies from low and middle-income countries that were published in languages other than English. Qualitative studies were also excluded from the current review, and may provide additional insight to quantitative, survey-based studies. Finally, we excluded a large number of studies and results that did not distinguish findings based on the type of child abuse. This was on the assumption that findings might differ between emotional abuse and other forms of child abuse, but meant that there were few studies examining some domains of interest.
This systematic review has revealed some key gaps in the literature. As noted above, we did not identify any studies that examined beliefs about causes of child emotional abuse or factors that are influential in its development, and few studies examined attitudes related to intervening or reporting emotional abuse. Further high-quality research, using random sampling methods, is required to gain a deeper understanding of the community’s knowledge and attitudes. This research should be tailored to the local setting, as knowledge and attitudes may vary depending on the cultural context and availability of support options. This research could then inform interventions to target incorrect knowledge or beliefs, change social norms about the acceptability of emotionally abusive parenting, and increase the likelihood of intervening or seeking support.
Conclusion
Child emotional abuse is a major public health problem. There is a need for more high-quality data on public understanding of child emotional abuse to inform educational strategies for improving its prevention and intervention.
Supplemental Material
Supplemental Material - Public Knowledge and Attitudes About the Nature and Impact of Child Emotional Abuse: A Systematic Review
Supplemental Material for Public Knowledge and Attitudes About the Nature and Impact of Child Emotional Abuse: A Systematic Review by Amy J. Morgan, Ellie Tsiamis, Cameron Tan, and Qiang Chen in Child Maltreatment
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
All data generated or analyzed during this study are included in this published article and its supplementary information files.
Supplemental Material
Supplemental material for this article is available online.
References
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.
