Child psychological/emotional abuse and neglect: A definitional conceptual framework

Abstract

Purpose

This study developed a definitional conceptual framework to clarify the nature of the definitions of psychological/emotional abuse and neglect (PEA) and to test the efficacy of this framework by applying it to three different types of definitions.

Methods

The definitional conceptual framework was developed from conceptual issues addressed by the previous literature. The framework was then applied to three types of definitions: (i) abstract definitions in the literature describing what PEA is, (ii) operational definitions in practice guidelines provided for professionals, and (iii) research studies on professionals’ definitions of PEA.

Results

The framework was able to show that definitions of PEA were often not specific about their inherent definitional components or how the ways in which they were similar or different from other definitions. To the extent that the nature of the definitions was explicit, the framework was able to clearly distinguish the considerable variation in the nature of PEA definitions both within and across the three types of definitions.

Conclusions

Policy, practice, and research on PEA (and other forms of child abuse and neglect) are unlikely to progress without clarity about the definitions being used (in terms of their conceptual components) and the rationale for using the specific definitions chosen. The framework provided by the study is an effective tool for describing the components used in PEA definitions and thus distinguishing the differences between definitions. The framework should also be applicable for other types of child abuse and neglect.

Introduction

The UN Convention on the Rights of the Child stipulates that children should be protected from all forms of maltreatment (United Nations, 1989). Psychological/emotional abuse and neglect (PEA) is one of the forms of child abuse and neglect (CAN). This study employs the term PEA to comprehensively encompass all relevant concepts (e.g. psychological maltreatment, emotional maltreatment, psychological abuse, emotional abuse, emotional neglect), involving both act of commission (abuse) and omission (neglect). However, when other specific terms are used in the literature, the original term is cited.

The academic literature contains a number of debates and suggestions about defining PEA (e.g. Baker, 2009; Garbarino et al.,1986; Glaser, 2002, 2011; McGee & Wolfe, 1991). Some argue that PEA is core to (Hart & Brassard, 1987) or part of other forms of CAN (McGee & Wolfe, 1991). One of the earliest and well-known definitions by Garbarino et al. (1986) described PEA as a “concerted attack by an adult on a child’s development of self and social competence…” (p.8). The authors categorized PEA into five different forms of: “rejecting”, “isolating”, “terrorizing”, “ignoring”, and “corrupting” (ibid.). In contrast, Glaser (2002) defined PEA as a “care-child relationship that is characterized by patterns of harmful interactions, requiring no physical contact with the child. Motivation to harm the child is not necessary for the definition” (p.697).

There is no agreement about what are the defining features of PEA or an overarching conceptual framework to examine the nature of the variation in the definitions. Defining PEA is problematic and its difficulty has been expressed by leading scholars. Glaser (2011) stated that “one of the main difficulties regarding emotional abuse has been the issue of defining it” (p.867). Similarly, Iwaniec (1995) wrote that:

The literature (although not all that extensive) is full of contradictions. Some difficulties lie in the absence of unified and precise definition of what exactly constitutes emotional abuse, and in how it is possible to provide measurable evidence that would be convincing and scientifically sound. (p. 3)

The lack of clear conceptualization of PEA could hinder the development of research, policy, and practice, and prevent society from developing clear strategies to protect children from PEA. Therefore, this study reports on the use of a structured approach based on a definitional conceptual framework to specify inherent components of different types of definitions of PEA.

Objective

Previously, the literature did address or discuss some of the conceptual components involved in definitions of PEA or CAN broadly (e.g. Baker, 2009; Garbarino et al., 1986; Glaser, 2002, 2011; Gough, 1996b; Iwaniec, 1995; McGee & Wolfe, 1991; O’Hagan, 1993, 1995). However, these were not always examined comprehensively or consistently. The objective of this study is to further evolve the conceptual understanding of the definitions of PEA. To achieve this, a definitional conceptual framework was developed to examine the clarity and consistency of three types of definitions of PEA: (i) abstract definitions of what PEA is, (ii) operational definitions for identifying and responding to PEA in guidelines for professional staff, and (iii) research on how professionals understand PEA (in both general terms and how they operationalize these).

The research questions addressed in this study are:

• How do abstract, operational, and research definitions of PEA vary?

• Is the definitional conceptual framework feasible and helpful in explaining the nature of PEA definitions?

Methods

For this study, a definitional conceptual framework was developed to include eight conceptual components based on the following conceptual mechanism which identified in the existing literature (e.g. Baker, 2009; Belsky, 1991; Brassard et al., 1987; Garrison, 1987; Garbarino et al., 1986; Glaser, 2002, 2011; Gough, 1996a, 1996b; Hart & Brassard, 1987; Hutchison, 1990; Iwaniec, 1995; McGee & Wolfe, 1991; O’Hagan, 1993, 1995).

For abuse and neglect to occur, there must be someone or something causing the abuse and neglect. This could be an individual, a group, an organization, or a whole society. Definitions of abuse and neglect may not specify who or what this is, but a source of the abuse and neglect is logically present. Abuse and neglect also imply that there is some action or lack of action or behavior that is inappropriate or results in abuse and neglect. This behavior may vary in its frequency from single to multiple events. The abuse and neglect may be an intended or a not specifically intended outcome of the behavior. It may also be part of an interaction between the abuser and the abused and lead to physical, mental or social consequences for the child. For abuse and neglect to occur, there must also be the abused child or children who may be defined by age and other characteristics.

Definitions of abuse and neglect, including PEA, thus involve at least the following eight conceptual components: abuser characteristics, abusive behavior, frequency, intention, consequences, interaction, child age, and other child’s characteristics. These components are all interrelated and not independent of each other. There may be other aspects of abuse and neglect that could be involved in definitions, but these eight components are core logical components providing a core definitional framework for PEA.

In this current study, this conceptual framework was applied to three types of definitions of PEA: abstract definitions, operational definitions, and research on professionals’ definitions. The aim was to examine the nature of the variation in definitions and to assess the effectiveness of the conceptual framework in such an analysis.

Analysis (i): Abstract definitions

The first analysis investigated general definitions of PEA (i.e. general description of what PEA means) based on the academic and other relevant literature. These definitions are developed and used for various purposes in diverse contexts to provide a general understanding of what PEA is. As the definitions are not necessarily associated with any specific practical actions, they are described in this paper as ‘abstract definitions’. This is in contrast to the ‘operational definitions’ of PEA discussed later in the paper.

A purposive search strategy was used to identify and examine different types of abstract definitions of PEA. Four different sources were searched: academic literature; primary research studies; major international or national/federal organizations; and definitions provided by NGO, NPO, or charities. To be included in this study, the definitions had to be statements which describe the meaning of PEA. A process of saturation was used so that when definitions were identified that were the same as already included definitions, these were not further included in the analysis.

A total of 34 abstract definitions were identified in the academic and other relevant literature (see Table 1). These definitions were examined using the definitional components to grasp the nature of and variation in the definitions of PEA. First, the definitions were investigated as to how they did or did not refer to each of the eight conceptual components. Second, a cluster analysis was undertaken to examine how definitions were grouped in terms of whether they referred to different combinations of the conceptual components.

Table 1.

A list of references of abstract definitions

Authors/source (Alphabetical order)	Type of resources
American Humane Association (n.d.)	NGO, NPO or charities
Australian Institute of Family Studies (2012)	International organizations or national/state/federal government
Barnett et al. (1991)	Original study
Brassard et al. (1993)	Academic literature on definitions
ChildLine (n.d.)	NGO, NPO or charities
Doyle (1997)	Original study
Garbarino et al. (1986)	Academic literature on definitions
Gavin (2011)	Academic literature on definitions
Glaser (2011)	Academic literature on definitions
Hibbard et al. (2012)	Academic literature on definitions
Ike (2006)	Academic literature on definitions
International Conference on Psychological Abuse of Children and Youth (1983)	Academic literature on definitions (The definition is cited in Hart & Brassard (1987))
Iwaniec (1995)	Academic literature on definitions
Kairys et al. (2002)	Academic literature on definitions
Leeb et al. (2008) (Centers for Disease Control and Prevention)	International organizations or national/state/federal government
Martin & Walters (1982)	Original study
McGee & Wolfe (1991)	Academic literature on definitions
Ministry of Community Development, Youth and Sports (2005)	International organizations or national/state/federal government
MHLW (n.d.)	International organizations or national/state/federal government
Moran et al. (2002)	Original study
Morelen & Shaffer (2012)	Academic literature on definitions
NSPCC (n.d.)	NGO, NPO or charities
O’Hagan (1993) (emotional abuse)	Academic literature on definitions
O’Hagan (1993) (psychological abuse)	Academic literature on definitions
Ontario Association of Children’s Aid Societies (2009)	NGO, NPO or charities
Prevent Child Abuse America (n.d.)	NGO, NPO or charities
Prevent Child Abuse North Dakota (n.d.)	NGO, NPO or charities
Schmitt (1978)	Academic literature on definitions
Skuse (1989)	Academic literature on definitions
UN Secretary General (n.d.)	International organizations or national/state/federal government
UNICEF Innocenti Research Centre (2005)	International organizations or national/state/federal government
WHO (2002)	International organizations or national/state/federal government
WHO (2007)	International organizations or national/state/federal government
WHO & ISPCAN (2006)	International organizations or national/state/federal government

Note: O’Hagan (1993) argued to separately define ‘emotional abuse’ and ‘psychological abuse’ and proposed two definitions in the literature, and the analysis included both definitions.

Individual components of definitions

The degree to which the 34 definitions considered each of the eight definitional components was examined.

(i) Abuser characteristics. The occurrence of PEA implies that there is someone or something that is responsible for that negative situation, but not all definitions refer to who or what this abuser might be.

The majority, four-fifths of definitions (80%; n = 27), referred to an abuser but in different forms. More than half of the definitions (65%; n = 22) limited the meaning of abuser by using terms such as carer (or caregiver, caretaker) or parent. For these definitions abuser means someone in a specific relationship to the child. Glaser (2011), for example, refers to the abuser as someone who regularly takes care of the child. Other definitions (15%; n = 5) left the meaning of abuser broad and vague with terms such as “a person” (Kairys et al., 2002), “adult” (Garbarino et al. 1986) and “perpetrator” (Gavin, 2011; Moran et al., 2002). These definitions acknowledge that there is someone responsible for PEA but do not specify in detail who this can be and so imply that even a stranger, who might not have any relationship or caring responsibility to the child can potentially be a PEA abuser. Interestingly, none of the definitions consider abuser characteristics beyond the individual level such as group, organization, or society.

Types of abuser characteristics referred to in definitions:

• Abuser not overtly referred to (20%)

• Abuser referred to – narrow definition (65%)

• Abuser referred to – broad definition (15%)

(ii) Abusive behavior. PEA also implies an active or lack of behavior by an abuser to an abused. Almost all the definitions (97%; n = 33), except one (Martin & Walters, 1982), referred to abusive behaviors. There are a number of different forms of behaviors described in the definitions. Garbarino et al. (1986) categorized PEA into five types of behaviors -rejecting, isolating, terrorizing, ignoring, and corrupting. This is an influential definition cited by many in the literature on PEA (e.g. Brassard et al., 1987; Hamarman & Bernet, 2000; McGee & Wolfe, 1991; Moran et al., 2000). However, some definitions also included behaviors, such as blaming (e.g. Gavin, 2011; WHO, 2007), degrading (e.g. Moran et al., 2002; Leeb et al., 2008), discrimination (e.g. ChildLine, n.d.; WHO, 2002) and exposure to violence (e.g. NSPCC, n.d.; Ontario Association of Children’s Aid Societies, 2009), which did not fit exactly into the five categories of Gabarino et al. (1986). This may imply that the scope of abusive behaviors has expanded over time.

Rejection was one of the most common sub-categories of abusive behavior identified in the abstract definitions. In some definitions, there was further specification about what rejection can be. WHO (2007) and WHO and ISPCAN (2006) limits it to non-physical rejection. Garbarino et al. (1986) are similar in describing rejection as “the adult refuses to acknowledge the child’s worth and the legitimacy of the child’s needs” (p.8). On the other hand, Kairys et al. (2002) considered rejection as “avoiding or pushing away” (p.1), suggesting it could take a physical form.

PEA can be conceptualized as either an active (act of commission) or inactive (an act of omission) behavior. Explicitly or implicitly, the majority of definitions include both commission and omission (88%; n = 30). The rest of the definitions either did not refer to abusive behavior at all (2%; n = 1) or referred to abusive behavior in a vague way without specifying whether it was active or inactive behavior or both (8%; n = 3).

Types of abusive behaviors:

• Abusive behavior not overtly referred to (3%)

• Abusive behavior referred to – as both active and inactive behavior (88%)

• Abusive behavior – vague about active or inactive behavior (8%)

(iii) Intention. Definitions of PEA might or might not be underpinned by the abusers’ motivation to be abusive or neglectful. Most of the definitions did not refer to intention (74%; n = 25). A second group of definitions stipulated that intention was not a necessary component of PEA (15%; n = 5). Moran et al. (2002) argued that with or without abuser’s intention, a child is equally exposed to risks, therefore, they consider that the abuser’s intention is not an essential component of PEA. On the other hand, Ike (2006) acknowledges the clinical importance of the abuser’s intention but suggests defining PEA regardless of intention as it is so difficult to objectively judge the existence of intention. So these authors refer to intention but state that it is not a compulsory component for defining PEA.

The third group of definitions include intention as a necessary element of PEA (6%; n = 2). Garbarino et al. (1986) described PEA as “a concerted attack by an adult…” (p.8) which is planned behavior and so includes some sort of motivation and intention.

The fourth group refers to intention as a possible aspect of PEA (6%; n = 2). Gavin (2011) stated that “…aggression can also be indirect, taking the form of criticism, advice, or help, but with the intent to belittle and control the recipient” (p.57). Similarly, according to the NSPCC (n.d.), “emotional abuse can involve deliberately trying to scare or humiliate a child or isolating or ignoring them”. Both definitions refer to intention but do not specify whether it is a necessary component of PEA. Such ambiguity could allow flexibility but also result in confusion of interpretations about how to consider intention in relation to PEA.

It is important to note that intention could mean different things, for example, either intention to be abusive or to be harmful to the child. Some of the definitions in the current analysis gave examples of abuser intention but did not discuss or distinguished different types or meanings of intention.

Types of intention:

• Intention not overtly referred to (74%)

• Intention referred to – intention is not necessary (15%)

• Intention referred to – intention is necessary (6%)

• Intention referred to – in some examples of PEA (6%)

(iv) Frequency. An individual action or event could be defined as PEA or the definition might require some ongoing or repeated actions. More than half of the definitions (68%; n = 23) referred to the notion of frequency of action and many of these (47%; n = 16) referred to PEA as something that involved frequent behaviors. Other definitions referred to PEA as something that could be the results of either single or multiple actions over time (21%; n = 7).

Some definitions provided further information on frequency of PEA. One group of studies referred to a group of behaviors that combined to be abusive (24%; n = 8); for example, the combination of verbal abuse and over-expectations of child achievement. For single events being defined as abusive, some definitions (12%; n = 4) included the degree of severity in defining PEA. Ike (2006) and Ministry of Community Development, Youth and Sports (2005) only defined single events as PEA when the case was severe or extreme.

A further issue is events that take time to complete. Kairys et al. (2002) explained that contentious divorce as an example which could end up being PEA because this can be a very painful experience for the child which usually changes the structure of household and subsequent pattern of daily life. Divorce is an event, but the experience may continue over an extended time. In this way, a potential adverse effect on the child as a painful experience is due to the entire process rather than an event at a narrowly specified point of time.

Types of frequency:

• Frequency not overtly referred to (32%)

• Frequency referred to – defined as high frequency events (47%)

• Frequency referred to – defined as either a one-off or multiple events (21%)

(v) Interaction. PEA by individuals involves at least two people, the abuser and the abused. These people interact in a certain way for PEA to exist. Few of the definitions (11%; n = 4) addressed the idea of interaction (or type of communication or relationship) in the definition of PEA. Glaser (2011) has established a conceptual framework (FRAMEA) to understand PEA with five tiers of child-carer interactions. The author argued that PEA is persistent so that it eventually becomes a part of the relationship between the child and the abuser.

Types of interaction:

• Interaction not overly referred to (89%)

• Interaction referred to (11%)

(vi) Consequence. Children could experience consequences due to PEA. Most of the definitions (94%; n = 32) referred to negative impacts of PEA. This included psychological, developmental, and behavioral impacts on a child.

Definitions varied in the extent that consequence was perceived as a necessary integral part of PEA. More than half of the definitions (56%; n = 19) implied that consequence should be a component of PEA, whereas less than half (38%; n = 13) implied that consequence is an optional component of PEA. Ike (2006), for example, proposed that PEA could be defined solely on abusive behavior regardless of its consequences. Ike argues that it is not necessary to wait until negative consequences to appear to respond when early action could prevent or minimize the adverse effect on the child.

Types of consequence:

• Consequences not overtly referred to (6%)

• Consequences referred to– consequence is necessary (56%)

• Consequences referred to– consequence is an optional (38%)

(vii) Child age. Children differ in their ages and their developmental level. The definitions of PEA did not consider the age of the child except for one definition (3%; n = 1) that referred to child age in relation to differential adverse consequence for the child.

Types of child age:

• Child age not overtly referred to (97%)

• Child age referred to (3%)

(viii) Other child characteristics. Alongside child age, there might be certain characteristics of children that may be considered relevant to defining PEA. Characteristics such as difficult behaviors (Garbarino et al., 1986; Glaser, 2011), low IQ (McGee & Wolfe, 1991), and disabilities (O’Hagan, 1993) were considered as risks in making children vulnerable to PEA. However, none of the definitions specified any child characteristic as an integral part of defining PEA.

Combination of definitional components

The definitional components were investigated together to create an overall picture of PEA. A cluster analysis was undertaken based on the binary data using SPSS 24.

The specific criteria for the coding were:

• Abuser characteristics: (1) the definition of abuser includes parent, carer, or refers to family setting; (0) others (including when there is no reference about abuser).

• Abusive behavior: (1) abusive behavior has an implication of, or refers to, inclusion of active behavior and neglectful behavior; (0) others.

• Intention: (1) intention is considered as necessary criteria; (0) others (including when there is no reference to intentions).

• Frequency: (1) repetitiveness is considered as necessary to define PEA; (0) others (including when there is no reference to frequency).

• Interaction: (1) interaction is considered as necessary to define PEA; (0) others (including when there is no reference to interaction).

• Consequences: (1) there is reference to consequences; (0) there is no reference to consequences.

Child age and other child characteristics were excluded from the cluster analysis due to insufficient data.

The analysis identified three major clusters (Cluster A, B, C) of combinations of the definitional components making up the entire concept of PEA. Cluster A includes 13 definitions. All the definitions in this cluster refer to the conceptual components of abuser (as someone close to the child), abusive behavior (both active and inactive), frequency (as repetitive), and consequences of PEA. Cluster B includes 10 definitions, and all of these refer to abusive behavior and consequences. Also, the repetitive nature of the abuse (frequency) was referred to by the majority of the definitions (n = 8). The significant difference from Cluster A is that the majority (n = 9) did not specify the meaning of abuser as someone close to the child. Cluster C includes 11 definitions. Most of them included aspects of the abuser (n = 8), abusive behavior (n = 7), and consequences (n = 8). Unlike the two other clusters, the majority of definitions in Cluster C did not refer to PEA consisting of repetitive acts (frequency).

The characteristics distinguishing these clusters can be summarized as:

1. Only Cluster A demonstrated a clear pattern in combination of definitional components (i.e. abuser, abusive behavior, frequency, consequences).

2. The inclusion or absence of reference to the abuser (as someone close to the child) is a key distinction across the three groups. This suggests that the majority of Cluster A and Cluster C definitions limited the concept of PEA to interfamilial abuse or neglect. On the other hand, Cluster B conceptualized PEA more broadly including something which could happen outside of the family setting. This is a prominent distinction which affects the entire concept of the definition of PEA.

3. Abusive behavior and consequences were core components of all clusters. There is wide agreement that PEA involve active as well as neglectful behaviors.

The cluster analysis revealed these characteristics of PEA definitions, but it was unclear what underpins such differences. Thus, the association between these clusters and two characteristics of definitions (the types of literature resource and the year of publication – both shown in Table 1) was further explored, and it did not identify any association. This means that it is unknown whether the identified clusters reflect anything in particular or it is just that these definitions happen to be grouped based on what authors happened to define as PEA.

Overall, the analysis of the abstract definitions demonstrated the diverse nature of PEA within as well as across the definitional components. The application of the definitional framework to the abstract definitions has made explicit the conceptual differences between the definitions. The definitions do not provide a consistent specification of what PEA is.

Analysis (ii): Operational definitions

The second analysis examined operational definitions (i.e. how professionals are guided to operationalize the definitions of PEA), based on guidelines created for professional use. In other words, the focus of the analysis is on how professionals are advised to identify and respond to PEA in practice. This can include the threshold for when a situation becomes bad enough to require a professional response. It is important to note that professional guidelines can also include general abstract definitions of PEA but for this analysis the focus is on professional identification and response to PEA.

Depending on the social or organizational system, there are various ways to apply the definitions in practice (Gough, 1996a). Thus, guidelines from California, England, and Japan were included to give breadth to the analysis. Amongst these jurisdictions, there are commonalities as well as differences which make comparison meaningful. First, they are all part of the OECD, suggesting a reasonably common level of economic development. Second, they have relatively developed child welfare systems, though these differ in their details and historical background. Third, all of the jurisdictions indicate the presence of legislative or policy-based professional reporting responsibilities of CAN (Mathews & Kenny, 2008), suggesting there are some common strategies to tackle PEA.

The inclusion of these jurisdictions from different geographical areas allows comparisons that help understand the nature of how definitions are applied in practice. To be included in the analysis, guidelines had to be written for professionals and be national or state level on CAN/PEA (or regional level only if it is about PEA specific). Guidelines were searched through official websites of governments, Internet search engine Google, and by direct enquiry to government bodies via email. These guidelines may have changed since the data collection of this study, but the purpose is to show the nature of variation in the definitions applied rather than to be a statement of the current status of guidelines.

California

California has a state level guideline underpinned by the California Child Abuse and Neglect Reporting Act (CANRA) (Rady Children’s Hospital San Diego, 2012). CANRA was passed in 1980 and has been repeatedly amended. Under CANRA, mandatory reporters refer to a number of professional groups including educators, health professionals, social workers, firefighters, and clergy members. They all have a legal obligation to report PEA (as well as other forms of CAN) and failing to do so can in theory result in jail or a fine.

In the guideline, two thresholds of PEA were found. One was based on the legal requirement from CANRA to report PEA (legal threshold hereafter), and another was additional to CANRA in advising how to respond to PEA (non-legal threshold hereafter). Understanding the differences between these two aspects of the guidelines can illustrate the different thresholds of PEA being applied.

As be seen in Table 2, the legal threshold referred to the conceptual components of abuser characteristics, abusive behavior, intention, and consequence. However, the meaning of these components was not specified and therefore left ambiguous. In contrast, the non-legal threshold referred to components which were not included in the legal threshold (e.g. frequency) and also provided more concrete explanations of individual component than the legal threshold. Thus, the two thresholds differ in terms of the combination of definitional components as well as the level of explanation of each component.

Table 2.

Two thresholds of PEA in California

Definitional components	Legal threshold	Non-legal threshold
Abuser characteristics	“Any person”	“Caretaker”
Abusive behaviors	“cruelty”; “unjustified punishment”	“criticism”, “threats”, “rejection”
Intention	“Willful“	“intended to produce fear or guilt…”
Frequency	Not applicable.	“a pattern of behavior”; “consistent criticism…”
Interaction	Not applicable.	Not applicable.
Consequences	“mental suffering”	“Emotional and behavioral problems, in varying degrees…”, “attention deficits”
Child age	Not applicable.	Not applicable.
Other child characteristics	Not applicable.	Not applicable.

Moreover, the guideline provides some clarification about the boundary of the legal and non-legal reporting thresholds. First, the presence of adverse mental effects on the child (i.e. “mental suffering in and of itself”) does not on its own specify the threshold of legally reportable PEA but it does for the non-legal threshold. There is a statement that “mental suffering in and of itself” does not have to be reported and rather it ‘may’ be reported. Second, the guideline addresses the importance of how PEA comes to light, by stating that cases which meet the legal threshold often comes from verbal disclosure or direct observation.

The nature of PEA definition in the California guidelines:

• Two thresholds (legal and non-legal thresholds) differ in terms of the meaning of individual components as well as the combination of definitional components (see Table 2).

• The thresholds are described not only with the definitional components but also with other conditions (e.g. how a case comes to attention).

England

In England, where there is no mandatory reporting law, two types of guidelines were identified. One which addresses all types of CAN including PEA for the whole of England (CAN guidelines) (HM Government, 2015a; HM Government, 2015b; NICE, 2009), and another which is specifically on PEA in particular local geographic areas (PEA guidelines) (Bristol Safeguarding Children’s Board, n.d.; Nottingham City Local Safeguarding Children Board, 2014). These are all underpinned by the English national Children Act 1989 and Working together to safeguard children (HM Government, 2015a).

The central component of threshold for the CAN guidelines is the risk (consequences for the child) to require different levels of professional response (i.e. referral, immediate protection, child protection plan). The consideration of both potential as well as actual harm enables authorities to intervene even without explicit evidence of existing consequences as long as it is perceived that there is a risk of such harm.

The local PEA guidelines refer to the conceptual components of interaction, frequency, consequence on the child, the degree of consequence on the child (i.e. mild, moderate, serious), and links between parental behavior/status and state of the child. In regard to intention, there is a clear statement that this is not a necessary part of PEA. PEA guidelines were thus much more specific than general CAN guidelines in terms of application of the definition of PEA.

The nature of the definition of PEA in the English guidelines:

• The central definitional component defining the threshold of PEA is risk of significant harm (i.e. adverse consequence on the child) that may be assessed by information on carer behaviors.

Japan

In Japan, there is the Child Abuse Prevention Law (CAPL), which was established in 2000. In 2004, the meaning of reportable CAN, including PEA, was extended from ‘the child who was abused/neglected’ to ‘the child who might be abused/neglected’. There is no formal mandatory reporting law though suspected cases of CAN, just as in England, should be referred to designated agencies. CAPL is a fundamental law for all forms of CAN and explicitly includes PEA. Unlike England and California, there was no guideline produced for various professional staff in Japan. Thus, a guideline which covers child protection broadly (MHLW, 2013) and a guideline for a specific professional group (i.e. Yōgo teachers - “special licensed educators who support children’s growth and development through health education and health services” (Okada, 2011, p. 21)) (MEXT, n.d.) was used for the analysis.

The Japanese guidelines do provide explanation of PEA but are not explicit about the threshold criteria for PEA. One of the guidelines explicitly states that all forms of CAN should be judged based on the legal definition and also by taking account of other relevant information of the child and the family (MHLW, 2013). The legal definition of PEA by CAPL, which are referred in both guidelines, is predominantly described as some forms of abusive behaviors. This therefore implies that the threshold of PEA is defined by abusive behaviors.

In contrast, there are statements in the guidelines which imply that PEA is defined based on another aspect of PEA – the consequence for the child. The guideline emphasized the importance of child-centered approach in judging all forms of CAN and also stated that, regardless of the abuser’s intention, CAN needs to be judged based on whether the child rearing is ensured by being secure and healthy (ibid.). These statements imply consequences for the child are an important element to determine the threshold of PEA. From these, the threshold of PEA is ambiguous and there is even a contradiction of its criteria.

The nature of the definition of PEA in Japan guidelines:

• A rather ambiguous threshold for PEA with legal definition emphasizing abusive behavior and other aspects of the guidelines emphasizing consequences for the child.

Overall, the examination of operational definitions in the guidelines revealed an emphasis on different PEA conceptual components for identifying PEA across the three jurisdictions. California described the thresholds of PEA based on the degree of mandatory reporting responsibility. The legal threshold is designed to capture PEA cases that have strong evidence (e.g. consequences, intention), and the non-legal threshold determines the scope of PEA more broadly and could capture PEA cases that do not show obvious evidence, therefore, cannot be captured by the legal threshold.

Such different cut-off points of PEA enables differential responses, and it also allows monitoring of the different status of PEA cases. Mandatory reporting might lead to higher rates of reporting (Mathews & Kenny, 2008). If there are not sufficient resources (e.g. time, staff) then, “when mandatory reporting exists, the proportion of children investigated by child-protection services, who have maltreatment substantiated, is low” (Gilbert et al., 2009, p. 168). Such potential challenges of mandatory reporting law might be minimized by setting the two thresholds. Although it was unclear from the guidelines, the results might suggest that Japan, without mandatory reporting, has two levels with the legal emphasis on abusive behavior and a broader concern for child consequences. England also does not have mandatory reporting and with a single threshold of PEA based on consequence across different phases of child protection. Jurisdictions without mandatory reporting will, of course, still have legal criteria for removal of children at risk from their parents and also prosecution for crimes against children.

Regardless of the different definitions in the three jurisdictions, none of them provided detailed criteria based on the eight conceptual components for responding to possible PEA. This may mean that professionals have to rely more on their own knowledge, experience, and intuition in their professional practice.

Analysis (iii): Research on professionals’ definitions

The third and final analysis was based on research studies that examine how professional practitioners define PEA. The literature was searched through major databases (PsycINFO and PubMed) and Google Scholar using the combination of relevant concepts (PEA, professionals, perception, decision-making). The studies had to meet the criteria of: being a primary study; including a professional staff as study participants; and examining the professionals’ views or/and attitudes about PEA.

A common method of studying professionals’ definitions of PEA was the use of vignettes describing scenarios that might or might not be considered PEA. There were studies based on other methods, but the next section only discusses the findings of some examples of vignette studies. The use of conceptual components of PEA in these studies is then considered.

Vignette studies of definitions of PEA

Comparing PEA to other forms of CAN. Four decades ago, Giovannoni and Becerra (1979) used vignettes to examine whether different groups of professionals agreed about the level of seriousness of certain acts of maltreatment based on vignettes. The study found that PEA was regarded as less serious than other types of CAN, suggesting the high threshold of professional abstract definitions of PEA. Other studies have also reported similar results (Collier et al., 1999; Pierce & Bozalek, 2004).

Studies of PEA alone. Baker et al. (2021) conducted a vignette survey of the views of professionals in various disciplines of child protection in the US of 18 parental behaviors, which were preselected by experts as behaviors representing PEA. Most respondents considered 4 of the 18 behaviors (threatens suicide; says “I hate you”; ignores distress; untreated child’s suicide) to be PEA, and more than half of the respondents indicated that their definite intent to report regarding 4 behaviors (threatens suicide; untreated child’s suicide; ignores pain; untreated swollen ankle). The study also showed that most respondents anticipated negative rather than positive outcomes as a result of PEA.

Burnett (1993) undertook a vignette study on PEA and reported that situations involving repetitive PEA with negative consequences were considered more serious than single episodes of PEA. However, as negative consequences and frequency were not manipulated separately within the vignettes, it is unclear which component affected the responses. The high rating could have been due to the nature of high frequency, negative consequence, or both.

Baily and Baily (1986) conducted a study to develop, refine, and rate the definitions of PEA. The study involved a total of 207 respondents from different protective service professions in the US. The respondents identified different definitions by the professionals for different age groups of preschool (ages 0–5), latency (ages 6–12), and adolescents (ages 13–17), suggesting that child age influences the people’s perceptions regarding PEA. The authors also reported that the majority of respondents (85%) stated that abusive behaviors alone could define PEA.

Different respondents’ definitions of PEA. In addition to the nature of a case, respondents’ characteristics can also influence the nature of PEA definitions. Burnett (1993) reported that female respondents were more likely to consider vignettes as abusive and serious than male respondents. Also, Fraser et al. (2010) found that older nurses and those with a parental role had a lower threshold for defining PEA.

Another potential difference is the nature and level of professional responsibility that people have towards PEA. Pierce and Bozalek (2004) reported that social workers were less likely to consider PEA vignettes as serious compared to other professional groups and lay people. This may imply that the less professional responsibility people have, the more they consider a situation to be serious. On the other hand, there were also studies which did not identify any difference between different groups. Burnett (1993) did not find a difference between social workers and citizens, and similarly, Carleton (2006) did not find difference between mandatory and non-mandatory reporters.

From these findings, it can be that the characteristics of both the PEA cases and of the professionals could affect these professionals’ abstract definitions. They are abstract definitions as the respondents are being asked whether something is an instance of PEA rather than about thresholds for responding to possible PEA.

Abstract versus operational definitions. There are also studies asking respondents how they would respond to different presented vignette scenarios. These studies find that just as with the abstract definitions, the professionals studied had a higher threshold for operational definitions of PEA compared to other forms of CAN (Beck et al., 1994; Shor, 1997; Shor, 1998).

There is some evidence illustrating a gap between professionals’ abstract and operational definitions. Shanalingigwa (2010) undertook a vignette study amongst three groups of US born parents, African born parents, and child welfare professionals. The study found that 34.9% of professionals defined “ignoring the child” as abusive but only 19% said that they would report it. The professionals’ responses indicated that they would not report all of the presented situations as PEA even if they thought they were examples of PEA.

Conceptual components in vignette studies

The discussion of the results of vignette studies shows that they can be a powerful way of studying definitions of PEA. However, the constructs of PEA used in the vignettes varied.

First, the description of individual definitional components differed. In general, vignettes were constructed predominantly based on abusive behaviors. However, different numbers and forms of abusive behaviors were identified across the studies. Some studies examined a greater number of abusive behaviors than others. Further looking at the nature of abusive behaviors, the boundary of PEA and non-PEA was treated differently. For example, Carleton (2006), in a study specifically on PEA, included vignettes which could be considered as sexually abusive behavior (i.e. encourages child to flaunt sexual charms).

Second, the inclusion of combination of definitional components varied. Vignettes in Giovannoni and Becerra (1979) involved abusive behaviors and consequences. In contrast, vignettes of Burnett (1993) included abusive behaviors, frequency, and consequence.

The diverse constructs and lack of specification of definitional components in the vignettes in the studies makes it difficult to aggregate and synthesize study findings. In summary, the lack of consistency and comprehensiveness of the specification of the conceptual components of PEA in the research design has limited the ability of the studies to progress and accumulate our knowledge of professionals’ abstract and operational definitions of PEA.

Discussion and conclusions

This paper reports on the development of a conceptual framework of PEA and its use to systematically examine three types of definitions of PEA: abstract definitions, operational definitions, and the definitions used in studies of how professionals define PEA. The study has shown that there is considerable variation and lack of clarity about how PEA is understood and defined within and across the three types of definitions.

PEA definitions from theory to practice

This study added evidence on the definitions of PEA, from theory to practice. Figure 1 below illustrates the relation between the three analyses. Analysis (i) added evidence on theoretical aspect of PEA by investigating the nature of abstract definitions in the relevant literature. Then, Analysis (ii) examined the operational definitions in the professional guidelines. Subsequently, Analysis (iii) examined professional practice of PEA, that is, how professionals perceive and respond to PEA, based on research studies.

Fig. 1.

Three analyses of PEA definitions from theory to practice

The results revealed the diverse nature of definition of PEA. Individually, the definitions did not clearly specify how they differed from other definitions both within and across the eight definitional components. In addition, these definitions were not clear about the criteria for defining situations as abusive or neglectful for the conceptual components that they did reference. There was therefore considerable variation in what was meant as PEA between and within the three different types of definitions.

The significance of the definitional conceptual framework

PEA definitions vary in both the nature of the definitional components applied and in the criteria and ‘cut off’ for deciding the degree of behaviour or consequence is sufficient to merit the definition of abuse or a particular practice response. A lack of detailed specification of definitions may have advantages in allowing flexible interpretations of children’s situations depending on context.

Regardless of such potential advantage of obscurity, for policy, practice and research on PEA to progress, there needs to be clarity about what PEA is. An individual specific definition of PEA may seem logical and make sense. However, the systematic unpacking the PEA definitions into their component parts has shown the extensive diversity of the nature of the definitions. These differences in definitions have practical consequences for the development of policy, practice guidance and the specification of research samples. The conceptual framework developed for this study provides a way to more clearly specify definitions and to enable discussion on what definitions would be appropriate and why in different circumstances.

The framework has been effective in clarifying the differences in abstract definitions and thus can provide a foundation for enabling further discussions about what is the PEA that we as individuals and society are concerned about.

The necessity for greater conceptual clarity also applies to operational definitions. To provide services to support families and for the state to use powers to intervene in family life requires specification and criteria for thresholds of the response of such services and interventions. Fine-tuned judgements may require case-by-case interpretation and judgement by professionals’ agencies and the courts, but the general parameters about what is acceptable childcare should be clear. The current study has shown, however, that operational practice guidelines are not precise about these issues.

The conceptual framework provides a means by which operational definitions can be proposed and professionally and democratically discussed and developed. The framework can assist case management by clarifying what is known and unknown about individual cases. It can also lead to more detailed improved management data about the nature of cases and service provision. The current study did not examine how in practice definitions of PEA are applied in day to day practice so this requires further investigation.

Conceptual clarity of definitions is also essential for research studies. When research is designed and undertaken without clarifying the nature of the phenomena they wish to study, then it hinders the credibility of research findings and prevents the accumulation of evidence across different studies.

Overall, this paper argues that greater clarity of the definitions of PEA is crucial for developing societal awareness, operational practice, and research on PEA. It has demonstrated the efficacy of a conceptual framework for making key conceptual components of definitions of PEA transparent. Although the focus of this paper has been on PEA, the conceptual framework and the arguments for conceptual clarification and a componential analysis of definitions is relevant to and can be applied to all types of CAN.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Availability of data and material

The author confirms that the data supporting the findings of this study are available within the article.

Code Availability (software application or custom code)

Not applicable

Declarations

Conflicts of interest/Competing interests

The author declares that there is no conflict of interest.