Child-Reported Family Violence: A Systematic Review of Available Instruments

Abstract

The impact of family violence (FV) on children is a significant global public policy issue. Earliest identification of FV among children is critical for preventing escalating sequelae. While practitioners routinely ask adults about FV, there are relatively few measures that enable children to reliably self-report on their own safety. This review sought to systematically identify and appraise all available child self-report measures for screening and assessment of FV in both clinical and research settings. Database searching was conducted in January 2022. Articles were eligible for review if they included a validated child (5–18 years) self-report measure of FV (including victimization, perpetration, and/or exposure to inter-parental violence). Screening of an initial 4,714 records identified a total of 85 articles, representing 32 unique validated instruments. Results provide an up-to-date catalog of child self-report measures of FV, intended to benefit practitioners, services and researchers in selecting appropriate tools, and in understanding their suitability and limitations for different cohorts and practice goals. While just under half of the measures captured both exposure to inter-parental violence and direct victimization, none captured all three domains of exposure, victimization and perpetration together. Instruments with provision for input from multiple respondents (e.g., both child and parent report) and with assessment of contextual risk factors were few. Findings point to the need for developmentally appropriate, whole-of-family screening and assessment frameworks to support children in the early identification of family safety concerns.

Introduction

Including Children in Family Violence Screening and Assessment

Children are rarely included in family violence (FV) screening and assessment. Earliest identification of FV through valid screening tools and structured enquiry with children is critical for reducing both acute and chronic harm (Hornor, 2005). Screening is an early identification strategy for use in settings such as antenatal, mental health, child and family health, and family law services. Assessment is a more in-depth, intervention-focused approach designed to support children and families where screening has identified FV or where there is a noted risk of FV. While family services increasingly ask adults about FV (McIntosh et al., 2016), there are comparatively few clinically relevant tools that enable children to reliably self-report on their own perspective of safety in the family (Edleson et al., 2007; Nygren et al., 2004; Saxton et al., 2020). Children’s experiences of FV may include exposure to inter-parental violence and/or direct maltreatment, which may be physical, sexual or emotional.

Detecting FV can be challenging even for well-trained practitioners in high-risk settings, for example, family law and mediation settings. Evidence from adult populations indicates that around 50% of FV remains undetected in mediation services when there is no routine, behaviorally specific FV screening (Ballard et al., 2011). Among children, detection of FV can be even more difficult. For example, there is evidence from clinical cohorts that children’s reports of parent-child aggression are significantly lower than parent reports of the aggression on the same measure (Jouriles et al., 1997). Children and young people may be unwilling to disclose violence in the family due to a number of reasons. Where there is structured clinical screening for FV with a practitioner, the process of screening may disrupt a child’s sense of comfort with or trust in the practitioner (Loveday et al., 2022), or children may be confused about the purpose of the questions. Low reporting rates of FV experiences by children can also stem from disempowerment or trauma-related symptoms that interfere with recall of information (Graham-Bermann et al., 2006) and loyalty to or fear of a perpetrating family member (Goodman-Brown et al., 2003). However, without developmentally specific and structured ways of collecting information about FV from the child’s perspective, clinicians and researchers could risk missing important information. It is therefore important for clinicians and researchers to have thorough, reliable and developmentally appropriate ways of gathering child reports of FV.

Despite the potential sensitivities that need to be considered when gathering child-reported FV data, restricting screening to the perspective of only one family member—often a parent when this occurs in family service settings—can obscure nuance in and magnitude of risk. For example, evidence from a large, representative household sample of 12 to 18-year-olds (n = 1,093) and their parents has shown that overall levels of agreement between parent and child reporting on maltreatment are low to moderate (Chan, 2012). In this sample, parents of both genders tended to underreport severe physical violence against their children compared to children, but were more likely to report minor physical violence, psychological aggression, and neglect (Chan, 2012). Similarly, in a recent scoping review of child maltreatment reporting practices in general population surveys (Stewart-Tufescu et al., 2022), where studies involved multi-informant reports, the congruence between informant- (e.g., caregiver) reported and child-reported maltreatment experiences was low to moderate. In these research survey samples, children and youth reported more maltreatment than caregiver respondents. Evidently, agreement between child and parent reports may vary in different ways across different settings. In both clinical and research settings, similar challenges regarding child disclosure could occur, for example, challenges related to rapport and trust (with a clinician or an interviewer) and confusion about questions. While some validated instruments may be suitable for collecting information about FV from children in research settings, it is important to understand the suitability of these instruments in clinical settings where information may then be used to inform support or intervention.

While parents and children often report divergent perspectives on FV, research on children’s lived experiences of FV is minimal (Noble-Carr et al., 2020) and few screening and assessment approaches are developmentally tailored to children (Buckley et al., 2007; Chan, 2012). Whole-of-family approaches that integrate the experiences of multiple family members within a single instrument or toolkit are rarer still (McIntosh et al., 2016).

Screening and Assessment Approaches

In healthcare settings, population- or cohort-based screening refers to a test offered to everyone in a target group, to improve health outcomes and reduce the burden of disease by way of identifying early signs of illness. Where FV is concerned, systematic screening for FV exposure is critical for best practice in “indicated” populations (Braaf & Sneddon, 2007; Johnston & Ver Steegh, 2013; Stanley & Humphreys, 2014). In this case, indicated populations are target cohorts that are known to be at increased risk of FV (e.g., in family law settings) or cohorts for whom the outcomes of FV are particularly risky (e.g., in antenatal care settings). “Universal” screening approaches refer to screening of the entire population attending services where the risk of FV is pronounced, for example, a child and family services center or a mediation service.

Here, screening for FV is delineated from assessment of FV impact. Certain instruments are suitable only for screening, while others can be used to support more in-depth assessment. The current review aims to identify both. Screening typically plays a triage function with the aim of differentiating individuals or groups who require further support based on responses to the screen, from lower-risk individuals who do not require further support. When risks are identified via screening, assessment can follow, through extended enquiry into circumstances, context, experiences, and risk status. Assessment approaches enable elaboration of risk, which usually has an intervention purpose such as making tailored clinical referrals.

Importantly, certain screening approaches may have limited utility in certain settings. For example, there is evidence that screening is not always fruitful for detecting exposure to inter-parental violence in healthcare settings (O’Doherty et al., 2015). While there is robust evidence that adverse childhood experiences (ACEs) including maltreatment and exposure to FV are associated with increased risk of poor health outcomes across the population (Kalmakis & Chandler, 2015), more research is needed on the utility of ACEs screening, including when, where, and how to screen, and which ACEs to screen for in which settings (Cibralic et al., 2022; Finkelhor, 2018; Meehan et al., 2022). Results from recent systematic reviews (Cibralic et al., 2022; Loveday et al., 2022) suggest that screening for ACEs can increase the identification of adversity which may in turn increase referrals to services, but there are limited data to indicate whether this then results in increased referral uptake or improved mental health outcomes. There are recent arguments that general ACEs screening is not appropriate for routine or widespread clinical use, as this approach may fail to accurately identify those at highest risk and fail to lead to the most appropriate intervention (Cibralic et al., 2022; Finkelhor, 2018). As such, more research is needed to understand the benefits of screening for certain types of ACEs such as those related to FV. While evidence supports screening for FV as a critical first step in the identification of harm, there is further work to be done in identifying which instruments should be applied to particular cohorts and settings (Meehan et al., 2022), and what gaps exist in the availability of such instruments. This is needed to inform refinements in FV identification and intervention policy in indicated settings, with consideration given to any known barriers for families in taking up referrals.

FV and Child Maltreatment

Child self-report of exposure to FV, usually inter-parental violence, is relatively less developed than child self-report of other forms of maltreatment, such as emotional maltreatment, neglect, physical and sexual abuse (Laurin et al., 2018). All can result in actual or potential harm to child development, health and wellbeing (Ferrara et al., 2015), with the potential for lifespan (Messman-Moore & Bhuptani, 2017) and intergenerational impacts on socio-emotional well-being and mental health (McDonnell & Valentino, 2016). Children are harmed not only through direct maltreatment but also in the course of exposure to violence in the home, including being raised by a parent who has experienced intimate partner violence (McIntosh et al., 2021). Children who witness violence between their parents have similar outcomes to those who experience direct victimization; exposure to inter-parental violence is associated with significantly worse psychosocial and neurocognitive developmental outcomes in children as compared to non-exposure to violence (McIntosh et al., 2016). Global prevalence comparisons are made difficult by limited standardization in definitions, data collection and reporting of maltreatment across countries (Ferrara et al., 2015). Conservative estimates suggest that 1 in 15 of all children under the age of 18 experience maltreatment, including FV exposures (Svevo-Cianci et al., 2010).

When identifying FV experiences among children and youth, a consideration of violence perpetration by the child is also warranted given that young people who are victimized may also display unsafe behaviors toward others (Chow et al., 2022). In detecting violence or maltreatment within the family system, child perpetration of violence may provide a clinically useful indicator of patterns of violent behavior at the broader family level. It is also important to screen for child perpetration behaviors in order to detect when targeted early intervention or support for both the child and their family may be needed.

The Importance of Contextual Factors

With regards to FV, contextual factors refer to events or circumstances that commonly co-occur with FV or increase the likelihood or magnitude of FV. Different forms of FV, including victimization, perpetration and exposure to inter-parental violence usually co-occur for children within the context of family and contextual stressors (Herrenkohl et al., 2008; Lee et al., 2021; McGuigan & Pratt, 2001), with compounding negative outcomes (Herrenkohl et al., 2008; Messman-Moore & Bhuptani, 2017). For example, FV risks are pronounced in the presence of stressful life events including poverty and financial strain (Bassuk et al., 2006), and particularly so with specific events such as conflicted parental separation or divorce (McIntosh et al., 2016; Wells et al., 2018). Other family stressors including financial strain, parental unemployment, parenting stress, mental illness and substance abuse are all associated with risk for child exposure to FV (Herrenkohl et al., 2008). Surrounding risk factors such as community or neighborhood disadvantage are also relevant to safety within the family system (Herrenkohl et al., 2008). Including questionnaire items that detect important contextual factors related to FV can improve specificity in screening and may aid precision in providing clinical and social support when families are facing multiple challenges, including contextual factors that may increase the risk of or enhance the magnitude of FV. However, questionnaire items related to contextual factors are not always included in screening or assessment measures of FV, and therefore being able to easily identify which measures include the detection of contextual risk is likely to be useful for clinicians and researchers in a range of settings.

Child-Reported Violence in Research Settings

While the primary rationale of this review is to identify existing instruments for child report of FV based on the observation that service settings (e.g., antenatal, mental health, child and family health, and family law services) do not frequently include children in FV screening and assessment, research settings also require suitable, validated measures for child reported experiences of violence in the family. Use of self-reported evidence on maltreatment by children, rather than reliance on evidence reported only by parents or other respondents, is the gold standard for research purposes. Direct, prospective reporting by children and youth in research studies circumvents issues related to reporting or recall bias when surveys are used retrospectively and/or by respondents other than the child (Stewart-Tufescu et al., 2022).

Given the current paper makes use of a systematic review methodology to identify available instruments, the instruments identified will naturally be drawn from primarily research articles. As such, some of these instruments will be suitable for research settings and some for both research and clinical settings. Therefore, a consideration of applicability of instruments to different settings is a core focus of the current review, and findings may be of benefit to both practitioners and researchers.

Aims and Scope

Practitioners and researchers need to be able to draw on reliable, validated measures that detect FV in a safe, evidence based and clinically useful way, yet recent and replicated evidence about child specific instrumentation is lacking (Edleson et al., 2007; Laurin et al., 2018). Our aim was to synthesize the existing evidence to address the question: What validated screening and assessment measures for FV account for the unique experience of the child via self-report? We do so via a systematic review of the availability, validity, and utility of FV instruments for report by children, encompassing exposure to FV, direct victimization, and also perpetration of violent behavior, in published studies from the past 20 years (2002–2022).

Method

Review Methodology

This review followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Page et al., 2021) over the following phases: (a) refining the research question and appropriate search terms; (b) utilizing the PRISMA guidelines (Page et al., 2021) to systematically search relevant databases; (c) using inclusion and exclusion criteria to systematically screen retrieved studies; (d) undertaking data extraction; and (e) narrative synthesis of findings. The review protocol was developed and registered on PROSPERO (CRD42022339693).

Search Strategy

A pilot search was conducted in January 2022 in four databases (Medline, Scopus, Cumulative Index of Nursing and Allied Health Literature [CINAHL], and PsycINFO) to test search terms and calibrate the data extraction table. The pilot search identified a total of 14,434 records. Preliminary review of pilot results for key articles identified relevant search results in CINAHL and PsycINFO with many relevant results not appearing in Medline or Scopus. Thus, to align with aims and scope, only CINAHL and PsycINFO were systematically searched (January 2022). Search results were restricted to peer reviewed articles published between 2002 and 2022 to ensure a contemporary review of the literature.

Search terms relating to risk screening (“risk screen*” or “risk assess*” or “safety screen*” or “safety assess*”), children (child* or kid* or offspring or adolescen* or preadolescen*) and FV (FV or domestic violence or abuse or domestic abuse or victim* or physical abuse or intimate partner violence or parent-child violence) with Medical Subject Headings (MeSH)headings were used. Reference lists of included review articles were hand-searched to identify additional relevant articles.

Study Eligibility

All records were screened for eligibility via (a) title, keyword, and abstract screening; and (b) full-text record screening. Studies were included if they (a) included self-reported data by a child or adolescent aged 5 to 18 years, (b) used a validated measure with available psychometric data, and (c) measured experience of FV . This included child perpetration to a member of the immediate family, physical, sexual or emotional abuse perpetrated by a member of the immediate family, or exposure to inter-parental violence (IPV) between parents or parent figures. Eligible study designs included experimental, randomized controlled trials, systematic reviews, mixed methods or qualitative studies (where the sample was aged between 5 and 18 years old).

Exclusion criteria included (a) data that were not self-reported by a child or adolescent, (b) studies that reported data on child’s emotional or behavioral problems but not FV, (c) studies that focused on violence, bullying or abuse external to the family (e.g., school bullying or community violence), (d) articles in which the FV instrument was part of a population-level government survey or census that was not suitable for FV screening or assessment specifically (for a review of these types of surveys, see Campeau et al., 2022), (e) unpublished literature (e.g., dissertations) and (f) articles in a language other than English.

Study Screening and Selection

The systematic search identified a total of 4,714 records. Following removal of duplicates in Endnote and Covidence, three reviewers (ZG, AV, and HVD) commenced title and abstract screening applying the inclusion and exclusion criteria to the remaining 4,230 records. Three reviewers (ZG, HVD, and AB) independently double-screened 20% of records at the title and abstract level (n = 846). Inter-rater agreement was moderate to high, with proportionate agreement ranging from 87% to 91% (Cohen’s κ = .56–.63). Conflicts were resolved via conferencing with a third reviewer (AB). A single reviewer (either ZG, AV, or HVD) screened the remaining 80% of records at title and abstract level. This resulted in 449 studies for full text review. One review author (ZG) then completed an in-depth full text review and excluded 335 studies. Of those excluded, 148 (44% of 335) included the wrong study design (i.e., did not use child self-report), 98 (29% of 335) reported on outcomes outside of the study scope (e.g., community violence or peer violence), 66 (20% of 335) included the wrong sample population (i.e., age range of participants did not meet the inclusion criteria) and three (1% of 335) were not available in English. The review management software Covidence (Veritas Health Innovation, 2017) was used to support the screening process. See Figure 1 for a PRISMA diagram outlining the identification, screening, exclusion and inclusion process of identified records (Page et al., 2021).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-analyses chart showing article identification process.

Data Extraction, Synthesis, and Analysis

Data were extracted by two reviewers (ZG, AV) using a standard data extraction form. For all included studies, data extraction was limited to set data items, which consisted of author (year), instrument name (including measure citation and psychometric features), country of origin for the measure, number of items, sample size, participant age (mean and standard deviation), category of FV captured by the measure (exposure, victimization, and/or perpetration), reporting of multiple-informant FV data (e.g., whether parent-report or clinician-report was used in addition to child self-report), and reporting of contextual factors (i.e., contextual characteristics relevant to the child’s experience).

Two reviewers (AB, ZG) further synthesized the data according to developmental level (i.e., measures suitable for children [5–12 years] and/or adolescents [12–18 years]), utility of the measure for universal screening or indicated assessment, and frequency of FV measure citation to determine the most commonly cited measures.

One reviewer (AV) extracted information about the purpose or setting of each instrument (i.e., whether the instrument was suitable for clinical or research purposes). This information was gathered from the identified records as well as from a subsequent directed search of any related psychometric papers or clinical guidelines pertaining to each instrument.

Results

Following full-text review of identified articles, 131 studies met the established inclusion criteria for data extraction: that is, they included a child self-report measure of FV. Of the 131 articles meeting inclusion criteria for review, 85 reported on previously validated child self-report measures of FV (32 unique validated measures; see Table 1). The remaining 46 articles reported on study-devised questionnaires or collections of items that lacked independently reported reliability and validity psychometrics. These are not included in narrative synthesis. Implications for practice, policy, and research and critical findings are presented in Tables 2 and 3.

Table 1.

Overview of Child and Adolescent Self-Report Measures of Family Violence.

Measure Name, Citation, Country	Description	No. of Items	Universal Screening (S)/Indicated Assessment (A)	Clinical (C) and/or Research (R) Application b	Type of FV			Dev. Period		Multi-Respondent/Alternative Versions	Contextual/Ancillary Factors	No. of Times Used
					Exp.	Vict.	Perp.	5–11 years	12–18 years
Frequently cited (>5 times)
1. Childhood Trauma Questionnaire—Short Form (CTQ-SF) Bernstein and Fink (1998) United States of America	Self-report inventory that screens for history of abuse and neglect across five subscales: physical abuse, emotional abuse, sexual abuse, physical neglect, and emotional neglect. Each subscale is composed of five items rated on a five-point Likert scale.	28	S	C; R Validated in clinical sample with children and/or adolescents.		X			X	No	Yes	17
2. Juvenile Victimization Questionnaire, 2nd Revision (JVQ-R2) Hamby et al. (2011) England	Screens for specified victimization types that cover five general areas of concern: conventional crime, child maltreatment, peer and sibling victimization, sexual victimization, and witnessing and indirect victimization. Asks “Have you suffered the following harms in your family in the past year?.” When the respondent answers “yes” to a victimization screener, they are asked follow-up questions regarding timing, incident, perpetrator and victim characteristics.	30+ (depending on no. of follow up questions)	S and A	C; R Validated in clinical sample with children and/or adolescents.	X	X			X	Parent-report version available (used mostly for children under 10 years)	Yes	14
3. Parent-Child Conflict Tactics Scale Straus (1979), Straus et al. (1998) United States of America	Measures psychological and physical maltreatment and neglect of children by parents, as well as nonviolent modes of discipline. Can be used with children as respondents, wherein each item can be asked about the mother and the father. The standard scale was designed to reflect three broad categories of conflict management tactics: Nonviolent Discipline (4 items), Psychological Aggression (5 items), Physical Assault (13 items) and Neglect (5 items).	22 (plus 14 supplemental questions if needed)	S	C; R Validated in clinical sample with children and/or adolescents.		X	X	X	X	Parent report version available	No	9
4. Kiddie-Sads-Present and Lifetime Version (K-SADS-PL)—PTSD section Kaufman et al. (1996) United States of America	Structured interview administered by a clinical psychologist. Includes a traumatic event sub-scale which is used to record exposure to traumatic events throughout the lifetime. Traumatic events listed in this subscale include car accidents, other serious accidents, fire, witness of a natural disaster, witness of a violent crime, confronted with traumatic news, witness to domestic violence, physical abuse, and sexual abuse.Childhood trauma defined as a history of physical abuse, sexual abuse, witnessing domestic violence or any combination of these.	25-item screener, plus follow-up questions when indicated	S and A	C; R Validated in clinical sample with children and/or adolescents.	X	X		X	X	Parent report version available	Yes	6
Occasionally cited (2–5 times)
5. (Montefiore) Child Clinical Adverse Childhood Experiences Questionnaire Murphy et al. (2014), Dube et al. (2001), Felitti et al. (1998) United States of America	10-category self-report questionnaire that measures lifetime exposure to trauma-related events, including experiencing a natural disaster, being involved in a car accident, witnessing domestic or neighborhood violence, and experiencing neglect, or emotional abuse. It also includes dimensions of household and familial dysfunction, including substance abuse in the home, divorce, or separation, and having a family member incarcerated.	19	S	C; R Evidence of use in clinical setting.	X	X			X	No	Yes	5
6. Violence Exposure–Revised (VEX-R), home violence questions Fox and Leavitt (1995) United States of America	Administered using a cartoon format to assess frequency of exposure to violent and criminal events in the home. Children are shown cards depicting violent and criminal acts and are asked to describe the frequency of their exposure to those acts, either as a victim or witness. Measure yields separate witnessing violence and violence victimization scale scores.	23	S	C; R Evidence of use in clinical setting.	X	X		X	X	Parent report version available	Yes	4
7. Maltreatment and Abuse Chronology of Exposure (MACE) Scale Teicher and Parigger (2015), Isele et al. (2014) United States of America	Developed to gauge severity of exposure to ten types of maltreatment during each year of childhood. Contains categories that directly represent physical abuse, physical neglect, emotional neglect, and sexual victimization. Emotionally abusive experiences, however, are reflected by four scales named Parental Verbal Abuse, Parental Nonverbal Emotional Abuse, Witnessing Violence Between Parents, and Witnessing Violence Toward Siblings.	52	S	C; R Evidence of use in clinical setting.	X	X			X	No	No	2
8. Survey of Children’s Exposure to Community Violence Richters and Saltzman (1990) United States of America	All measures of violence exposure reflect lifetime exposure, and include both witnessing and directly experiencing (i.e., victimization) violence: Witnessing violence at home, witnessing violence at school, witnessing violence in the community, and directly experiencing violence in the community (victimization). Witnessing violence at home measured with four items. Original items did not include context-related wording because the instrument was designed to measure exposure to community violence. To measure witnessing violence at home the words “in your home” were added to these items.	50 (4-items specific to violence in the home)	S	Not reported.	X				X	No	No	2
9. The University of California at Los Angeles Post-traumatic Stress Disorder Reaction Index (UCLA-PTSD-RI) Steinberg et al. (2004), Rolon-Arroyo et al. (2020) United States of America	Self-report instrument for assessing PTSD symptoms in school-aged children and adolescents. Questions are rated on a 5-point Likert-type scale from 0 (none of the time) to 4 (most of the time) based on the past month. Includes assessment of traumatic events including witnessing domestic/family violence and direct victimization such as emotional or psychological maltreatment.	22	S	C; R Validated in clinical sample with children and/or adolescents.	X	X		X	X	Child, adolescent and parent-report versions available	Yes	2
10. Stressful Life Events Schedule (SLES) Child-Reported Version. Williamson et al. (2003) United States of America	One SLES variable indicates exposure to domestic violence and sexual abuse (unclear whether sexual abuse was specific to FV). Subjects rate subjectively how stressful an event is on a four-point scale.	61	S	C; R Validated in clinical sample with children and/or adolescents.	X	X			X	Child, adolescent and parent-report versions available	Yes	2
Cited once
11. Adolescent Health Review (AHR) Harrison et al. (2001) United States of America	Designed for use in primary care settings, consisting of three demographic items (age, gender, race/ethnicity), a school enrolment item, and 29 items related to 14 risk domains: (a) lack of exercise, (b) poor nutrition, (c) unhealthy weight control, (d) family interaction problems (perception of parents as caring, family having fun), (e) problems at school, (f) emotional distress, (g) suicidal behavior, (h) violent behavior, (i) sexual activity, (i), cigarette smoking, (j) alcohol use, (k) Marijuana use, (l) Substance abuse, (m) Physical or sexual abuse (physical abuse by an adult in the family; sexual abuse by an older or stronger person, date rape or other date violence).	32	S	C; R Validated in clinical sample with children and/or adolescents.		X			X	No	Yes	1
12. Adverse Childhood Experiences Questionnaire (ACE-IQ) WHO (2018) International a	10 ACEs included in the World Health Organization ACE international questionnaire (ACE-IQ): sexual, physical, or emotional abuse; emotional neglect; parental substance abuse; parental mental illness or suicide attempt; violence between parents; parental separation; bullying; and parental criminal conviction.	43	S	C; R Validated in clinical sample with individuals aged 18 years and over.	X	X		X	X	No	Yes	1
13. Assessment of Liability and EXposure to Substance use and Antisocial behavior (ALEXSA) Ridenour et al. (2009) United States of America	Illustration-based, computerized child report assessment for early manifestations and predictors of substance abuse and antisocial behavior prior to high school. Includes a subscale on family conflict: Count of the dysfunctional tactics used to resolve family disagreements, for example, “When people in your family disagree, do they ever call each other names?”	246	S	C; R Validated in clinical sample with children and/or adolescents.	X			X	X	No	Yes	1
14. Behavioral Health Screen (BHS) Diamond et al. (2010) United States of America	Composed of 13 modules: demographics, medical, school, family, safety, substance use, sexuality, nutrition and eating, anxiety, depression, suicidal risk, psychosis, trauma and abuse. The BHS contains four items regarding whether the young person (a) had been physically or sexually hurt by a romantic partner, (b) had been physically or sexually hurt by an adult in their home, or (c) had a family member to confide in.	55 (plus 41 additional items when relevant core items are endorsed)	S	C; R Validated in clinical sample with children and/or adolescents.		X			X	No	Yes	1
15. Berkeley Puppet Interview (BPI) Ablow and Measelle (1993) United States of America	An interactive age-appropriate interviewing technique eliciting self-perceptions from 3.5 to 8-year-olds. The BPI currently consists of a number of separate domains and scales: (a) the BPI Family Environment Scales assess children’s appraisals of central family relationships (parent-child, marital, sibling, inter-parental conflict/fighting) and family processes (e.g., marital conflict, self-blame for conflict, shared-nonshared environment, perceived parental rejection); (b) the BPI Academic (BPI-A) and Social scales (BPI-Soc) tap children’s perceptions of teacher, school, and peers; (c) the BPI Symptomatology Scales (BPI-S) assess children’s perceived symptomatology in clinically relevant domains.	58	S	C; R Validated in clinical sample with children and/or adolescents.	X			X		No	Yes	1
16. Child Exposure to Community Violence Checklist (CECV) Amaya-Jackson (1998) United States of America	Has five response options with scores ranging from 0 to 4, with higher scores indicating more frequent exposure. Certain items assess either direct victimization or witnessed violence, including domestic violence, physical and sexual abuse. An example item relating to FV includes: “Have you actually been beaten up by your family”	39	S	C; R Validated in clinical sample with children and/or adolescents.	X	X			X	No	No	1
17. Child Maltreatment Interview Schedule—Short Form Briere (1992) United States of America	Participants asked about the punishments that they have ever received from parents, stepparents, foster parents, or other adults who were in charge of the participant for at least 6 months, that would qualify as child maltreatment. for example, if they had been “pushed, spanked, grabbed, slapped or shoved,” “hit very hard,” “hit with an object,” “beaten or kicked,” “locked in a room for 5 hr or more or told you can’t have food for a whole day or longer,” “hurt by an adult in charge of you so that you were bruised, had broken bones, or were severely injured,” or “severely punished in some other way that we haven’t talked about.”	15 (plus additional follow up questions for endorsed items)	S	R Evidence of research use only.		X		X	X	No	Yes	1
18. Child-to-Parent Aggression Questionnaire (CPAQ) Calvete et al. (2013) Spain	Includes 20 items, 10 relating to the father and 10 relating to the mother. Within each block of 10 items, 7 describe psychological aggressions (e.g., insulting someone, taking money without permission, or threatening to hit), and three describe physical aggressions (e.g., hitting someone with something that could hurt or kicking). Adolescents responded by indicating how often they had performed each of the behaviors against their fathers or mothers, or against the caregivers who adopted the role of their parents (stepmother, stepfather, grandparents, etc.) Also includes open questions about the reasons for the aggressive acts.	10	S	C; R Validated in clinical sample with children and/or adolescents.			X		X	No	No	1
19. Community Experiences Questionnaire (CEQ) Schwartz and Proctor (2000), Richters and Saltzman (1990) United States of America	Evaluates exposure to violence as a direct victim and exposure to violence as a witness. Includes four subscales: witnessed violence within the community, witnessed violence within the home, victimization by violence within the community, victimization by violence within the home.	26	S	R Evidence of research use only.	X	X			X	No	Yes	1
20. Event History Calendar (EHC) Martyn et al. (2013) Martyn and Belli (2002) United States of America	Structured yet flexible assessment tool that facilitates recall of past events by utilizing past experiences as cues to remembering. This calendar approach can be adapted to any population and specialty to obtain pertinent information, which in this case is history of, or ongoing ACEs (such as physical, mental, sexual abuse; neglect; violence), in addition to risk behaviors and individual strengths.	Variable depending on context	S	C; R Evidence of use in clinical setting.		X		Not reported		No	Yes	1
21. Exposure to Violence Scale Orue and Calvete (2010) Spain	Measures direct and indirect exposure to violence in different contexts. Includes assessment direct victimization and witnessing physical and psychological violence in the home (e.g., “How often has somebody hit you at home?,” “How many times have you seen one of your parents assaulting the other one at home?”).	21	S	R Evidence of research use only.	X	X			X	No	No	1
22. International Society for the Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse Screening Tools, Children’s Home Version (ICAST-CH-C) Zolotor et al. (2009) International a	Measures victimization exposure across demographic items and childhood victimization items categorized into six forms: violence exposure, psychological abuse, psychological neglect, physical neglect, physical abuse, and sexual abuse. Respondents are asked to indicate perpetrators (adults, other children or both) and frequency (many times in the past year, sometimes in the past year, never or not in the past year but this has happened) for each of the childhood victimization items.	57	S	R Evidence of research use only. Other versions validated in clinical sample with children and/or adolescents.	X	X			X	Yes, parent-report available	Yes	1
23. Family Background Questionnaire (FBQ) McGee et al. (1997) Canada	Used to screen for physical abuse and witnessing domestic violence. Self-report version of the Record of Maltreatment Experiences-Revised and includes global severity ratings for multiple types of maltreatment experienced since childhood. The child physical abuse scale contains three items including: “. . .hit, kicked or punched you” and “. . .threw you against something.” The exposure to domestic violence scale contained four items such as “. . .beat up her/his partner” and “threatened her/his partner with a gun.” Ratings provided for both the primary maternal and paternal figure.	7	S	C; R Evidence of use in clinical setting.	X	X			X	No	No	1
24. Life Events Checklist for DSM-5 (LEC-5) Weathers et al. (2013) United States of America	Respondents indicate varying levels of exposure to each type of potentially traumatic event on a nominal scale including domestic violence (slapped, kicked, hit, bit, attacked or beaten up); sexual assault (someone touching our body or made to touch someone else’s body); and neglect (not having enough food, home alone or lack of care).	27	S	C; R Evidence of use in clinical setting.	X			X	X	No	Yes	1
25. Lifetime victimization and trauma history (LTVH) Widom et al. (2005) United States of America	Assesses lifetime trauma and victimization history through a structured in-person interview. Questions refer to “scary and upsetting things” that happen to people “at home, in their neighborhood, or someplace else” and cover seven categories of trauma (general traumas, physical assault/abuse, sexual assault/abuse, family/friend murdered or suicide, witnessed trauma to someone else, crime victimization, and kidnapped or stalked).	30	S and A	C; R Evidence of use in clinical setting.	X	X			X	No	Yes	1
26. LONGSCAN Witnessed violence/home & community (History of Witnessed Violence) LONGSCAN (2000) United States of America	Adolescent-report measure including seven dichotomous indicators of whether or not the child witnessed or experienced violence during their lifetime by any family member: (a) hit, kicked, slapped, or beaten up; (b) threatened with a knife; (c) threatened with a gun; (d) stabbed or cut with a weapon; (e) shot; (f) killed; or (g) sexually assaulted.	7	S	C; R Evidence of use in clinical setting.	X				X	No	No	1
27. Multicultural Events Schedule for Adolescents (M.E.S.A) Arizona State University (1997) United States of America	Measures whether children/adolescents in the home setting have been threatened by violence; experienced physical violence; theft of personal possessions; have been emotionally abused or neglected (e.g., being frequently shamed, ignored, or repeatedly told that you are “no good”).	5	S	C; R Validated in clinical sample with children and/or adolescents.	X	X			X	No	Yes	1
28. Multiplex Empirically Guided Inventory of Ecological Aggregates for Assessing Sexually Abusive Adolescents and Children (MEGA) Miccio-Fonseca (2009) United States of America	Assesses risk level for sexual improprieties and/or sexually abusive behavior in male and female youth (validated for ages 4 to 19), including youth with low intellectual functioning. Includes measurement of exposure to domestic violence and history of abuse (i.e., sexual abuse, physical abuse, and neglect/emotional abuse).	75	S and A	C; R Validated in clinical sample with children and/or adolescents.	X	X		X	X	No	Yes	1
29. Sibling aggression scale—adapted from the University of Illinois Bully Scale. Espelage and Holt (2001) United States of America	Respondents are asked how often in the past 30 days they have experienced or engaged in aggression perpetration in relation to their siblings and/or other children in their home.	6	S	R Evidence of research use only.			X	X		No	No	1
30. Things I Have Seen and Heard Questionnaire Richters et al. (1992) United States of America	Administered to children to assess types of violence both witnessed and directly experienced by children. Five items include violence witnessed in the home. Items are scored on a five-point Likert scale from 1 = never to 5 = many times.	15	S	C; R Evidence of use in clinical setting.	X			X		No	No	1
31. Youth Risk Behavior Survey Guedes and Lopes (2010) United States of America	Screens for general health risks and includes two questions that query lifetime experience of physical abuse: (a) “Does someone in your family hit you with slaps, punches and kicks?” and (b) “How often does your mother/father/step-mother/stepfather or other adult hit you?.” A question regarding inter-parental violence: “Has there been violence among adults in your family?”	87	S	C; R Evidence of use in clinical setting.	X	X			X	No	Yes	1
32. (Birmingham) Youth Violence Exposure measure Mrug et al. (2008) United States of America	Screens for exposure to violence both (witnessing and victimization). Endorsement of any item related to witnessing or victimization item is then followed by three contextual probes, asking whether the exposure occurred in the home, school, or neighborhood.	Up to 18-items (6 standard items plus 3 possible follow-up probes per question if endorsed)	S	R Evidence of research use only.	X	X		X	X	No	No	1

Note. The types of family violence measured in each instrument are indicated in bold text in the ‘Description’ column. C = suitable for clinical application, either as instrument has been validated in a clinical sample with children and/or adolescents or there is evidence of clinical application such as clinical guidelines; R = suitable for research application, as instrument has been used in a peer-reviewed research study.

^aACE-IQ and ICAST measures were developed by an international team of experts and have been field tested across the world. S = universal screening; A = indicated assessment.

^bThe information in this column was sourced from records identified via review as well as any related psychometric papers or clinical guidelines that were publicly available.

Table 2.

Implications for Practice, Policy, and Research.

This article provides an up-to-date catalog of currently available child self-report measures of family violence (FV), of benefit to practitioners and services in selecting appropriate tools and understanding their suitability for different cohorts and goals.

Findings suggest a need for the ongoing refinement of systemic and developmentally appropriate screening tools to identify FV risk in children. Specificity in screening is needed to detect family vulnerabilities and relevant contextual factors, and to aid precision in clinical support.

Further developments in the science and application of universal FV risk screening with dependent children will aid services to create a safe, enabling space for children to share their safety and security experiences within their family.

Table 3.

Critical Findings.

Systematic search of the peer reviewed literature from the past 20 years identified 32 unique, validated instruments available for screening and assessment of child reported experiences of victimization, perpetration and exposure to inter-parental violence.

While just under half of the identified instruments captured both exposure to family violence and direct victimization, none captured all three domains of exposure, victimization and perpetration together. Few instruments lend themselves to a family systems perspective that considers the views of multiple family respondents.

Instruments that accounted for contextual risk factors were scant. Currently, use of supplementary assessment instruments alongside screening of family safety remains necessary to account for contextual risk or protective factors.

Frequently Cited Instruments

The most frequently cited instruments in articles identified via review (>5 citations) were the Childhood Trauma Questionnaire-Short Form (CTQ-SF; Bernstein & Fink, 1998); the Juvenile Victimization Questionnaire (JVQ; Hamby et al., 2004), the Conflict Tactics Scale (CTS; Straus, 1979) and the Kiddie-Schedule for Affective Disorders and Schizophrenia-Present and Lifetime version (K-SADS-PL; Kaufman et al., 1996). Descriptive details of each can be found in Table 1.

Of these, both the JVQ and the K-SADS-PL measure exposure to IPV as well as direct victimization, and the CTS takes perpetration behavior into account as well as victimization. The CTQ-SF measures direct victimization only. Regarding alignment from multiple respondents, both the JVQ and the CTS have parent-report versions as well as child self-report; the CTQ-SF and the K-SADS-PL were child self-report only. Three of the four most frequently cited instruments (the CTQ-SF, the JVQ and the K-SADS-PL) account for contextual factors that may be relevant to the child’s experience of family safety. The CTS and the K-SADS-PL can be used with younger children (5–11-year age range) as well as with children over 12 years old.

Country of Origin

While some measures may have been adapted, translated, or validated for use in countries outside of where they originated, in our search the ACE-IQ (WHO, 2018) and ICAST (Zolotor et al., 2009) were the only measures that were originally designed to be used cross-culturally and internationally. Almost all other measures were developed and first validated in the United States of America (n = 26). The remaining measures were developed in Spain (n = 2), England (n = 1) and Canada (n = 1).

Length of Instruments

The number of items across the 32 measures ranged from 5-items (Multicultural Events Schedule for Adolescents (M.E.S.A.; Arizona State University, 1997) to 246-items (Assessment of Liability and EXposure to Substance use and Antisocial behavior; Ridenour et al., 2009), with a mean of 39-items, and a median of 27-items.

Purpose/Setting of Instruments

The instruments identified here have varied utility for clinical and research settings.

As shown in Table 2, 14 measures appear to have been validated in clinical samples with young children and/or adolescents (and were identified in this review because of their use in research settings). One measure has been validated in a clinical sample with individuals aged 18 years and over (the ACE-IQ).

A further 10 of the identified measures appear suitable for application in clinical settings (i.e., the measure has been used clinically, but no psychometric properties derived specifically from clinical samples were identified via our search of the related literature).

Six of the identified measures appear to be most suitable for research rather than clinical purposes. One of these was intended specifically for use by researchers only (the Child Maltreatment Interview Schedule—Short Form).

No information on purpose or setting could be found for one measure (the Survey of Children’s Exposure to Community Violence).

Universal Screening/Indicated Assessment

Most identified measures were designed as screening measures only (n = 28; 87.5% of 32) and four included items suitable for both screening and indicated assessment. These were the JVQ (Finkelhor et al., 2011), K-SADS-PL (Kaufman et al., 1996), the LTVH (Widom et al., 2005), and the MEGA (Miccio-Fonseca, 2009).

Types of FV Measured

Of the 32 unique validated measures identified: 24 measured direct victimization; 22 captured exposure to FV (e.g., witnessing IPV); and four captured risk of perpetration of FV (e.g., child-to-parent abuse/violence, or sibling abuse/violence). While just over half (n = 17) captured both exposure to FV and direct victimization, none captured all three domains of exposure, victimization and perpetration together. The most commonly cited measure for screening direct victimization was the CTQ-SF (Bernstein & Fink, 1998), which was used in 17 of the identified articles. The JVQ-R (Hamby et al., 2011) was used most commonly to capture child-reported FV across multiple domains, as it measured both victimization and exposure (n = 14). The Survey of Children’s Exposure to Violence (Richters & Saltzman, 1990) was one of the only measures which included FV items related to exposure only, and was cited twice. Regarding the measurement of perpetration only, this was captured by the Child-to-Parent Aggression Questionnaire (Calvete et al., 2013), which was cited once.

Developmental Period

Most questionnaires were suitable for use with adolescents aged between 12 and 18 years (n = 30). Only 13 have been used with samples of younger children (5–11 years old) and only three were specifically designed for use with children under 12 years old (all of which had only been cited once each). The most commonly cited measure that was suitable for both children and adolescents (5–18 years old) was the Parent-Child CTS (Straus et al., 1998), which had available versions for both age groups. The CTQ-SF (Bernstein & Fink, 1998) was the most cited measure specific for adolescents aged between 12 and 18 years old (n = 17).

Multiple Respondents

Seven measures identified in this review had versions that could be completed by multiple respondents. All were parent-report versions of the questionnaire. These were the JVQ (Hamby et al., 2004), the CTSPC (Straus, 1979), VEX-R (Fox & Leavitt, 1995), UCLA-PTSD-RI (Rolon-Arroyo et al., 2020; Steinberg et al., 2004), K-SADS-PL (Kaufman et al., 1996), SLES (Williamson et al., 2003), and the ICAST (Zolotor et al., 2009).

Contextual Factors

The majority of measures (n = 20) also included some appraisal of contextual factors, such as the recent death of a close friend/family member, parental separation, serious illness or injury, community violence, bullying, or problems at school. A small number of measures (n = 2) also included an open-ended question which allowed respondents to report “any other major upheaval” that had occurred in their family context.

Discussion

This article provides an up-to-date overview of child self-report measures of FV, intended to benefit practitioners, services and researchers in selecting appropriate tools, and understanding their suitability and limitations for different cohorts and practice goals. Systematic search of the peer reviewed literature from the past 20 years identified 32 unique, validated instruments used to screen, assess or measure child reported experiences of victimization, perpetration and exposure to IPV. The majority of instruments identified here appear appropriate for clinical use, either in light of having been validated in clinical samples with children and/or adolescents, and/or having evidence of clinical application such as the availability of clinical guidelines.

From articles initially identified, the vast majority of published self-report measures of FV overall were for adults. Among these, various instruments retrospectively measure adults’ childhood experiences of FV. There are also many instruments that focus on clinician reported FV risk to children. Instruments with provision for reporting by multiple family respondents (e.g., both child and parent report) are few, bolstering earlier calls to include both parent and child reports on FV in clinical screening and assessment frameworks (Chan, 2012; Compier-de Block et al., 2017) given divergent perspectives and under-reporting by parents.

We discovered that while just over half of the identified instruments for children capture both exposure to FV and direct victimization, none captured all three domains of exposure, victimization and perpetration together. Few instruments lend themselves to a family systems perspective that considers the views of multiple family respondents. Only seven measures identified here had aligned versions suitable for completion by parents. Given whole-of-family safety screening and assessment are important to a full appraisal of family safety in clinical and healthcare settings (Lee et al., 2021), particularly given the trend toward under-reporting of FV by children (Goodman-Brown et al., 2003; Graham-Bermann et al., 2006), this gap is notable.

Instruments that accounted for contextual risk factors were also scant. We note debate about this in the wider literature. For example, the Parent-Child CTS has been criticized for ignoring the circumstances under which FV occurs (Straus et al., 1990; Straus et al., 1998). However, the authors of this scale view the exclusion of contextual factors as a theoretically important strength of the scale, given its focus on measurement of behaviorally specific acts of violence. In clinical settings, appraising risk in context is key to response planning (Wells et al., 2018), and this gap too is notable. Currently, use of supplementary assessment instruments alongside screening of family safety remains necessary (Straus, 2017).

While other instruments have been adapted for use beyond their country of origin (e.g., tCTQ-SF has been translated into multiple languages—for a review see Georgieva et al., 2021), the scope of this review included research articles published in English. Only two instruments identified in the current search were specifically designed for worldwide use, primarily to allow for intra- and inter-country comparisons (the ICAST and the ACE-IQ). Most instruments have been developed in the United States of America, and most child abuse research to date has been conducted in affluent, western countries. While this may partly be because the current review excluded three articles that were not written in English, the overrepresentation of Western participants in research on violence against children is an inherent limitation, and the true scope of the problem may be unaccounted. In developing the ACE-IQ, the authors (WHO, 2018) recognized that many countries have yet to appreciate the major public health implications of FV. Data collected outside of westernized countries is needed to promote awareness and inform practice and policy (WHO, 2018), and local conversation with children is needed to reach child participants in a safe and culturally appropriate way. Given both the ICAST and the ACE-IQ have been developed for use in research only, there are few if any child self-report clinical screening tools for FV validated outside of western countries.

Strengths and Limitations of this Review

This review is the first, to our knowledge, to systematically identify and summarize all available child self-report measures of FV. This contemporary overview of available instruments will be of use to both researchers and clinicians and serves to identify strengths and gaps in currently available approaches for appraising child-reported experiences of FV. We have been able to comment on features of these instruments including their length, application to different settings, suitability for multiple respondents, scope with respect to type/s of FV measured, and appropriateness for different developmental levels. This catalog should support services, clinicians and researchers to select the most appropriate measures for their client/s or contexts. For example, clinical services may require instruments that include appraisal of contextual circumstances or are suitable for aligned reporting by multiple family members, while researchers may need instruments that provide a streamlined focus on one particular type of FV.

We did not include gray literature in the current review, and it is possible there are other available child self-report measures of FV not identified here. Many unpublished and study specific measures were identified but not included in the current synthesis. However, the use of unvalidated measures when collecting child reported FV is potentially problematic given the additional vulnerabilities of this population. In some circumstances, it may be difficult for a child or adolescent to assess the potential risks and benefits of providing their perspective on FV and therefore the quality, reliability and developmental appropriateness of the measure used to collect sensitive information from children or adolescents is paramount. Further, study specific or unpublished measures do not allow for direct comparison of child reported FV across studies or different countries. In the current review, restricting scope to peer reviewed, previously validated measures ensured a sharp focus on instruments that are suitable for wider use when working with this vulnerable age group.

Publication bias is a possible limitation of this review. The peer reviewed body of literature preferences articles with significant findings and particular strengths of effects, and it is possible that there are other relevant instruments used in studies that have remained unpublished. Our results are also limited by the inclusion of articles published only in English. It was outside scope of the current review to include articles in other languages.

We did not include a formal risk of bias or quality assessment in the current review as our purpose was to identify and summarize the characteristics of child self-report FV instruments, rather than to summarize the findings of the studies these instruments were used in. In lieu of a typical quality assessment framework, we have provided a critical appraisal of the included instruments themselves, which is more consistent with the aims of the current review.

Conclusions

Earliest identification of FV experiences among children is critical for reducing developmental and intergenerational harm. While family services increasingly ask adults about FV, and there is a well-established evidence base of research on adult-reported FV, there are comparatively few developmentally tailored clinical and research tools that enable children to reliably self-report on their own perspective of family safety.

The current catalog of instruments provides detail on the types of FV that can be screened for in available measures; the developmental stages these measures are appropriate for; whether measures include appraisal of contextual factors; and characteristics of measures including descriptions, length, and versions available. The 32 validated instruments identified here may present valuable additions to study designs, clinical toolkits or inform clinical implementation guidelines where family safety is concerned. Further attention must be next given to evidence-based application of instruments to particular settings, that is, in research or evaluation designs that explore the acceptability, feasibility and effectiveness of particular instruments in different service settings.

Findings of the current review suggest a need for the ongoing refinement of systemic and developmentally appropriate screening tools to identify risk. Specificity in screening is needed to detect family vulnerabilities and relevant contextual factors, and to aid precision in clinical support. Findings invite further consideration of various operational and training considerations: for example, how best to equip practitioners to screen and assess for FV among children with confidence and competence, and how and when to equip services or organizations to prioritize these approaches. Implementation considerations also include attention to whether FV screening represents a cost-benefit given the time and effort required of individuals and organizations to implement routine or regular screening, and if so, which type of screening and/or assessment is appropriate and useful for the given cohort. Further developments in the science and application of FV risk screening with dependent children will help to create a safe, enabling space for children to share their safety and security experiences within their family.

Author Biographies

Anna Booth is a Research Fellow in family psychology and family violence in the School of Psychology and Public Health, La Trobe University. Anna has a keen focus on research translation and works in partnership with government and community stakeholders to better understand and support the needs of vulnerable Australian children and families.

Zoe Guest is a Researcher in the School of Psychology and Public Health, La Trobe University. Zoe’s research examines family relationships, including issues related to family safety and the evaluation of early intervention parenting programs. Zoe is also a registered psychologist.

An Vuong holds a bachelor’s degree in psychological studies and is a Research Assistant in the School of Psychology and Public Health, La Trobe University. An has contributed to research projects that explore the mental health and socioemotional functioning of young children and families.

Henry von Doussa is a social researcher with key interests in family well-being and LGBTIQ+ health and well-being. Where the two intersect has, over the past decade, been the focus of his research. Mr von Doussa’s work also supports relational health in families where gender diversity is present.

Claire Ralfs is the CEO of Relationships Australia South Australia and the Australian Institute of Social Relations. An experienced family services educator and counselor with more than 20 years’ experience in the human services sectors, Dr Ralfs is acknowledged for her skill in designing service systems that improve client outcomes.

Jennifer McIntosh is a clinical and developmental psychologist, family therapist, and Director of The Bouverie Centre, La Trobe University. Professor McIntosh’s areas of expertise include the development of assessment methods to enable early triage to support children and families through recovery of trauma and resolution of conflict.