Abstract
Using a pooled data set of two waves of the National Surveys of Children’s Exposure to Violence, this study investigates links between indicators of socioeconomic resources and lifetime exposure to two different forms of child neglect (physical and supervisory), examines how neglect is associated with the risk of other types of victimization, and estimates the impact of neglect on trauma symptoms. Findings suggest that physical neglect is directly linked to economic stressors, while low parental education is consequential for both physical and supervisory neglect. Both types of neglect also were strongly associated with risk of other maltreatment and most other forms of victimization. Physical neglect was particularly strongly related to sexual abuse and witnessing sibling abuse, while supervisory neglect was most strongly related to risk for sexual victimization by a nonfamily adult. Although neglect is significantly associated with trauma symptoms, poly-victims had, by far, the highest levels of trauma symptoms.
Child neglect is the form of maltreatment most frequently reported to youth protection systems, with 75.3% of all children who received a child protection response in 2015 experiencing neglect (U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children Youth and Families, & Children’s Bureau, 2017). It was also the most common form of maltreatment found in the National Incidence Studies (NIS), designed to estimate the incidence of maltreatment in the United States independent of any involvement by child protection authorities; of the more than 1.25 million children experiencing maltreatment during the NIS-4 study year, 61% were neglected (Sedlak et al., 2010). Nevertheless, neglect remains the form of maltreatment that has received, by far, the least attention among researchers (Hildyard & Wolfe, 2002; Stoltenborgh, Bakermans-Kranenburg, Van Ijzendoorn, & Alink, 2013; Wolock & Horowitz, 1984).
Most research that has focused on child neglect has been based on children reported to child protection services (CPS). CPS data provide only a limited view of the problem because it inevitably captures only a fraction of neglect that is experienced by children in the United States and is subject to a variety of biases associated with identification and reporting of cases, most often capturing more severe cases with evidence of harm. A better understanding of child neglect in the full U.S. population of children and youth, the different forms it takes, and how it fits into broader maltreatment and victimization contexts require comprehensive national population-based studies. Yet, measuring and studying neglect in a general population survey is particularly challenging, despite the critical need to better understand this form of maltreatment (Clément, Bérubé, & Chamberland, 2016; Slack, Holl, Altenbernd, McDaniel, & Stevens, 2003). Neglect has been assessed in national population samples of children and adults (Finkelhor, Ormrod, Turner, & Hamby, 2005; Finkelhor, Turner, Ormrod, & Hamby, 2009; Turner, Finkelhor, Ormrod, & Hamby, 2010), but past studies have done little to differentiate types of neglect or show how they are related to the general context of child victimization.
Definition and Types of Child Neglect
One of the challenges is inconsistency in operationalizing and measuring neglect (Dubowitz et al., 2005). Unlike other forms of maltreatment, which are conceptualized as acts of commission, child neglect represents acts of omission, which are inherently difficult to specify and measure (Straus & Kaufman Kantor, 2005). After an extensive review of prominent neglect typologies developed by researchers, CPS classification systems, and most state child abuse and neglect statutes, Coohey (2003) defined three broad types of neglect that best represented the concepts: physical neglect, supervisory neglect, and emotional neglect. The current study focuses on the first two of these forms of neglect. Similar to Coohey’s definition, we operationally define physical neglect as when a parent or caregiver fails to provide the child with adequate medical care, nutrition, cleanliness/hygiene, and/or safe and sanitary living conditions (unlike Coohey’s definition, we do not include the provision of adequate education as a component of physical neglect). Also similar to Cooley, we define supervisory neglect as when a parent or caregiver fails to provide the child with adequate supervision to ensure protection from dangerous situations or people. Thus, we operationalize supervisory neglect as leaving children alone with no alternative caregiving arrangements, caregivers being present but too incapacitated to adequately care for the child, and/or allowing dangerous adults access to the child. Our conceptualizations of physical and supervisory neglect are also consistent with those of other researchers in the field (e.g., see Korbin & Krugman, 2014).
Socioeconomic Resources and Neglect
One important but controversial issue around child neglect concerns its strong association with socioeconomic status. In NIS-4, the rate of neglect for children in low socioeconomic status households was nearly 7 times higher than for other children (Sedlak et al., 2010). Thus, neglect may be unintentional, often linked directly to parent’s lack of financial or material resources (Slack, Holl, McDaniel, Yoo, & Bolger, 2004). At the same time, we know that, even among families living in poverty, most parents adequately care for their children. More information is needed on which components of socioeconomic resources or deficits, such as parental education, family income, and perceived financial strain, are most strongly linked to neglect and how such linkages may differ across different forms of neglect and other forms of maltreatment. Such information may begin to give us a more nuanced understanding of how and why socioeconomic-related deficits function to increase children’s risk for neglect, in particular, and provide insights concerning how interventions can be better focused to reduce specific forms of neglect.
Co-occurrence of Neglect and Other Victimizations: Accessing the Unique Impact of Neglect
It is also important to better understand how particular types of neglect may co-occur with other forms of victimization, both within and outside of the family context. Research has already documented substantial co-occurrence between neglect and other forms of maltreatment. For example, Mennen, Kim, Sang, and Trickett (2010) found that, among child welfare clients, 95% of neglected children also experienced some other form of maltreatment, with emotional abuse most often co-occurring with neglect. In a retrospective study of adults, those who experienced physical neglect in childhood were over 6 times more likely to have also experienced emotional abuse and almost 4 times more likely to have experienced physical abuse than those with no physical neglect (Dong et al., 2004). However, the overlap of neglect with victimizations other than direct maltreatment by a caregiver (e.g., peer physical assault, witnessing domestic violence, and dating violence) has not been addressed. For example, it is plausible that supervisory neglect, in particular, may be associated with increased risk of victimizations perpetrated by nonfamily members such as sexual assault by peers or nonrelated adults.
Considerable research has documented negative effects of neglect on child development and well-being, including substantially elevated internalizing and externalizing problems as well as cognitive deficits, both in the short and long term (Dubowitz, Papas, Black, & Starr, 2002; Hildyard & Wolfe, 2002; Spratt et al., 2012). Indeed, neglect is thought to inflict trauma and psychological harm because when children are left alone, unsupported, and unassisted in the face of challenges, they can experience some of the same distress and emotional dysregulation that comes along with threats of a physical and sexual nature (Perry, 2008).
However, the potential co-occurrence of neglect with other forms of maltreatment, and with forms of victimization occurring both inside and outside family contexts, makes assessment of the impact of neglect more complicated. For example, research has shown that children in homes with intimate partner violence (IPV) are 2–8 times more likely to experience neglect than those not exposed to IPV (Hamby, Finkelhor, Turner, & Ormrod, 2010; Zolotor, Theodore, Coyne-Beasley, & Runyan, 2007). Only a more comprehensive assessment of child victimization, one that includes not only other forms of maltreatment but also exposures across other domains of the child’s life, can allow for an examination of the unique and relative influence of child neglect. Given how common neglect appears to be, at least among children involved in CPS, it is important to better understand its individual effects on child well-being, both within the general population of children and youth and in the context of other potentially co-occurring victimizations.
The Significance of Poly-Victimization
Recent research on poly-victimization has clearly demonstrated the importance of measuring and accounting for multiple forms of victimization. Poly-victimization is the experience of multiple victimizations of different kinds such as sexual victimization, physical abuse, bullying, witnessing family violence, and exposure to community violence. Studies have demonstrated that a significant portion of children in the United States who identify as victims of sexual abuse or other single forms of violence are in fact poly-victims who have experienced many different types of victimization (Finkelhor, Ormrod, & Turner, 2007; Turner, Finkelhor, & Ormrod, 2010). Moreover, poly-victimization has particularly damaging emotional and behavioral consequences for children and youth (Cyr et al., 2013; Ford, Grasso, Hawke, & Chapman, 2013; Hickman et al., 2013), often accounting for the presumed effects of individual forms of victimization (Finkelhor et al., 2007; Turner, Finkelhor, & Ormrod, 2010).
In a large sample of violence-exposed children, Hickman et al. (2013) measured violence exposure in three different ways: total frequency of all lifetime exposure, total frequency of lifetime exposure by broad category (i.e., assault, maltreatment, sexual abuse, and witnessing violence), and poly-victimization, defined as exposure to multiple violence categories. The results indicated that only poly-victimization (exposure to two or more categories of violence exposure) emerged as a consistent predictor of negative symptoms (Post-Traumatic Stress Disorder and behavior problems). In a study of youth involved in the juvenile justice system, investigators conducted latent class analysis derived from 19 types of adversity, including different forms of violence exposure. Among the three unique classes of youth that emerged, the poly-victims group was distinct from other adversity groups in the likelihood of exposure to multiple types of traumatic victimization and in the presence of severe emotional and behavioral problems (Ford et al., 2013). Using latent class analysis on a nationally representative sample, Turner, Shattuck, Finkelhor, and Hamby (2016) found six unique victimization profiles. The poly-victim group was more likely than the other classes to have been victimized in multiple settings by multiple perpetrators, experienced the most serious victimization characteristics (e.g., incidents involving a weapon, injury, or sexual a component), and had the highest trauma scores. Several additional studies further document the particularly strong links between poly-victimization and negative outcomes in children and adolescents (Cyr et al., 2013; Elliott, Alexander, Pierce, Aspelmeier, & Richmond, 2009; Ford, Elhai, Connor, & Frueh, 2010; Holt, Finkelhor, & Kaufman Kantor, 2007; Soler, Kirchner, Paretilla, & Forns, 2013).
Taken together, evidence strongly suggests that poly-victimization—that is, exposure to multiple different forms of victimization—is a powerful predictor of negative child outcomes that may account for many individual forms of victimization. Given the particular significance of poly-victimization, considering the potential link between different types of neglect and poly-victimization in assessments of impact is clearly important.
The objective of this study is to expand the available epidemiology on neglect using a national population survey and situate it better in terms of its component types, its connection to general family deprivation, and its interrelationship to other forms of maltreatment and to the broader contexts of child victimization. In light of these potential linkages, this study then seeks to determine the independent and relative influence of child neglect on trauma symptoms. Using a nationally representative sample of children and youth in the United States, the specific aims of this study are to: Determine how exposure to physical and supervisory neglect may be differentially associated with various components of parental socioeconomic resources including parental education, family income, perceived financial strain, use of public assistance, and parental unemployment history and whether such associations are similar to or different from those evident for other forms of maltreatment; Determine the relative risk of exposure to other forms of maltreatment as well as other forms of nonmaltreatment victimization, given exposure to physical and supervisory neglect; and Assess the impact of lifetime exposure to neglect on trauma symptoms, independent of exposure to other forms of maltreatment, nonmaltreatment victimization, and poly-victimization.
Method
Data
This study uses the two most recent waves (2011 and 2014) of the National Surveys of Children’s Exposure to Violence (NatSCEV) which collected data on wide range of victimization experiences including neglect, other forms of maltreatment (physical, emotional, and sexual abuse), and other types of victimization (physical assault, sexual assault, witnessing violence, peer victimization, dating violence, and conventional crime) using the Juvenile Victimization Questionnaire (JVQ). The data are representative of children and youth in the United States. The sample size of the aggregated data set was 7,852 children for youth ages 2–17 years. Sample characteristics are shown in Appendix A.
Procedure
A short interview was conducted with an adult caregiver (usually a parent) in each household to obtain family demographic information. One child was randomly selected from all eligible children living in a household by selecting the child with the most recent birthday. If the selected child was 10–17 years old, the main telephone interview was conducted with the child. If the selected child was under age 10, the interview was conducted with the caregiver who “is most familiar with the child’s daily routine and experiences.” A safety protocol was implemented to ensure confidentiality of responses and privacy during the interview.
Respondents were promised complete confidentiality and were paid US$20 for their participation. The interviews, averaging 45 min in length, were conducted in both English and Spanish. Two hundred and seventy-nine of the interviews with the parents were done in Spanish. Nearly all of the adolescents aged 10–17 chose to be interviewed in English. Respondents who disclosed a situation of serious threat or ongoing victimization were recontacted by a clinical member of the research team trained in telephone crisis counseling whose responsibility was to stay in contact with the respondent until the situation was resolved. All procedures were authorized by the institutional review board of the University of New Hampshire.
Sampling
For each of the two waves of data collection, a sampling frame was constructed using four sources: (1) an address-based sample of households from which cell phone and residential numbers could be dialed, (2) a prescreened sample of households with children from recent national random-digit-dialed surveys, (3) a listed landline sample (with a known child in the household based on commercial lists), and (4) cell phone numbers drawn from a targeted random-digit-dialed sample frame. This multistep frame construction helped to recruit households with children and to assure representativeness of the sample, paying special attention to include cell phone–only households.
For each wave, sample weights for all responses were applied to adjust for differential probability of selection due to (a) study design, (b) demographic variations in nonresponse, and (c) variations within household eligibility.
Measures
Neglect was measured through the lifetime endorsement of 6 items in the JVQ, 3 items that assessed components of physical neglect (care neglect, unsafe home neglect, hygiene neglect) and 3 items that assessed components of supervisory neglect (parental incapacitation, parental absence, inappropriate adults in the home).
Any physical neglect was measured as exposure to one or more of the following. Care neglect: “When someone is neglected, it means that the grown-ups in their life didn’t take care of them the way they should. They might not get them enough food, take them to the doctor when they are sick, or make sure they have a safe place to stay. At any time in (your child’s/your) life, was (your child/were you) neglected?” Unsafe home neglect: “Was there a time in (your child’s/your) life when (your child/you) lived in a home that was broken down, unsafe, or unhealthy. For example, it had broken stairs, toilets, or sinks that didn’t work, trash piled up, and things like that?” Hygiene neglect: “Was there a time in (your child’s/your) life when (his/her/your) parents did not care if (you were /he was/she was) clean, wore clean clothes, or brushed (your/his/her) teeth and hair?”
Any supervisory neglect was measured as exposure to one or more of the following. Neglect from parental incapacitation: “Was there a time in (your child’s/your) life that (your child/you) often had to look after yourself because a parent drank too much alcohol, took drugs, or wouldn’t get out of bed?” Neglect from parental absence: “Was there a time in your life when that (your child/you) often had to go looking for a parent because the parent left you alone, or with brothers and sisters, and that (your child/you) didn’t know where the parent was?” Neglect from inappropriate people in home: “Was there a time in that (your child’s/your) life when (your child’s/your) parents often have had people over at the house who (your child/you) were afraid to be around?”
Other forms of maltreatment
Although some researchers count exposure to IPV as a form of maltreatment (Gilbert et al., 2009), the NatSCEV conceptual framework considers only direct maltreatment (acts directed at the child), that is, physical abuse, emotional abuse, sexual abuse, and neglect, as forms of child maltreatment. Indirect types of victimization (not directed specifically at the child), such as witnessing IPV, are considered separate forms of victimization (included below). Consistent with most states definitions of maltreatment perpetrators (Child Welfare Information Gateway, 2013), we also consider only direct maltreatment incidents perpetrated by parents or other caregivers (i.e., babysitters, adult relatives) as maltreatment and incidents perpetrated by noncaregivers as other forms of victimization (included below).
Ten items from the JVQ were used to measure exposure to physical abuse (any physical assault or attempted assault by a parent, caregiver, or other adult family member), emotional abuse (child felt scared or really bad because parent or caregiver called him/her names, said mean things to him/her, or said they didn’t want him/her), and sexual abuse (any sexual assault or attempted assault by a parent, caregiver, or other adult family member).
Other types of victimization
Using questions from the JVQ, we measured eight other types of victimizations including witnessing IPV, witnessing sibling abuse, peer (nonsibling) physical assault, peer (nonsibling) sexual assault, physical assault by an adult (noncaregiver), sexual assault by an adult (noncaregiver), dating violence, and property crime. Base rates for all forms of victimization assessed in the current study are presented in Appendix A.
Poly-victimization was measured as total number of victimizations types to which children were exposed out of the 11 types assessed (all maltreatment and other victimization types listed above). Given our interest in co-occurrence with child neglect, the neglect items were not included in this count.
We note the JVQ, from which all the above measures were derived, has demonstrated good psychometric properties including test–retest reliability and construct validity (Finkelhor, Hamby, Ormrod, & Turner, 2005). Moreover, comparisons between proxy (i.e., parent) and self (i.e., child) reports with this instrument found no evidence of reporter bias (Finkelhor, Hamby et al., 2005; Finkelhor et al., 2009).
Socioeconomic resources was assessed with several indicators: parental education (highest level of parent/partner in household), total family income (measured on an ordinal scale using 10 categories and then recode into four income ranges), financial strain (the extent to which the parent agreed that there was not enough money for clothes or household items), currently receive any public assistance (do you currently receive Temporary Aid to Needy Families [TANF], food stamps, welfare, Medicaid, or any other public assistance?), and parental job loss (has there ever been times when (your child’s/your) mother, father, or guardian lost a job or couldn’t find work?).
Trauma symptoms includes anger, depression, anxiety, dissociation, and posttraumatic stress scales of two closely related measures: the Trauma Symptoms Checklist for Children (TSCC; Briere, 1996), which was used for the 10- to 17-year-old self-report interviews, and the Trauma Symptom Checklist for Young Children (TSCYC; Briere et al., 2001), used in the caregiver interviews for the 2- to 9-year-olds. The instruments were slightly shortened from their original forms for a total of 28 items in the TSCC and 25 items in the TSCYC. Respondents were asked to indicate how often they (or their child) had experienced each of the symptoms within the last month. Response options were on a 4-point scale from 1 (not at all) to 4 (very often). All item responses for the five scales together were summed and then converted to z scores to create an aggregate trauma symptom score. The TSCC and TSCYC have shown very good reliability and validity in both population-based and clinical samples (Briere, 1996; Briere et al., 2001).
Control variables
Demographic control variables included in multivariate analyses are child gender, age, race/ethnicity, family structure (single parent, two parent, parent with stepparent, and other, such as living with foster parent or grandparent), and household socioeconomic status (a standardized composite of household income and parent education).
Analysis
Chi-square tests were used to determine whether there were significant differences in rates of neglect by age and socioeconomic resources (Tables 1 and 2). Weighted logistic regression models were used to examine the relationship between neglect and other forms of victimizations, adjusting for age and socioeconomic status. Odds ratio were then converted to risk ratios using the Zhang and Yu (1998) methodology (Table 3). Weighted least squares regression models were used to determine the relationship between neglect and trauma symptoms (Table 4). The variables were entered in stages to determine the impact of each type of neglect independent of demographic controls (Models 1 and 2), the impact of each type of neglect relative to other types of maltreatment and other forms of victimization (Model 3), and the impact of each form of neglect relative to poly-victimization (Model 4).
Lifetime Prevalence of Exposure to Neglect.
Note. Asterisks indicate significant differences across age groups in exposure.
aAny endorsement of care neglect, unsafe home neglect, or hygiene neglect. bAny endorsement of parental incapacitation, parental absence, and inappropriate adults in home.
*p < .05. **p < .01. ***p < .001.
Socioeconomic Resources and Type of Neglect.
Note. Values in boldface are significantly different across socioeconomic resource groups at p < .05. TANF = Temporary Aid to Needy Families.
aAny endorsement of care neglect, home unsafe/unclean, or hygiene neglect.
bAny endorsement of parental incapacitation, parental absence, and inappropriate adults in home.
Adjusted Relative Risk Ratios for Neglect Types With Other Victimization Types (Weighted).
Note. Risk ratios calculated using Zhang and Yu (1998) methodology. Significance level determined from logistic regression results that adjusted for age and SES; additional analyses were conducted separately by age groups and the pattern of results in the two age groups was very similar, so age groups were combined to help increase power to detect associations involving lower base rates.
aAny endorsement of care neglect, home unsafe/unclean, or hygiene neglect. bAny endorsement of parental incapacitation, parental absence, and inappropriate adults in home.
*p < .05. **p < .01. ***p < .001.
Standardized Weighted Regression Coefficients Predicting Trauma Symptoms.
Note. All models are controlling for sociodemographic variables including child age, gender, SES, and race.
aPoly-victimization is measured as a count of the total number of victimization types experienced (excluding neglect).
*p < .05. **p < .01. ***p < .001.
Results
Socioeconomic Resources and Child Neglect
Table 1 displays lifetime rates of individual types of neglect, as well as composites of any physical neglect and any supervisory neglect, separately for the 2- to 9-year-olds (reported by a caregiver) and 10- to 17-year-olds (reported by the youth themselves). The rate of care neglect reported is significantly greater in the younger sample (5.27%) than the older youth sample (3.24%). The other types of physical neglect (unsafe home and hygiene neglect) do not differ by age-group. More striking are the much higher rates of supervisory neglect in the older age-group. This is particularly evident for parental absence, with over 9% of older youth endorsing this item, compared to a rate of only 1.6% for those under the age of 10. Although, relative to younger children, older youth have had a longer “opportunity” to have experienced supervisory neglect in their lifetimes, the lack of age difference in most forms of lifetime physical neglect suggests that supervisory neglect is influenced by the age itself. As discussed later, these age differences in supervisory neglect may reflect, in part, developmentally appropriate variations in supervision that may not always represent neglectful behavior on the part of parents. They may also be influenced by the different perspectives of parents and youth.
Table 2 presents associations between any physical neglect, any supervisory neglect, and family socioeconomic resources or deficits. Children of parents with less than a high school education had twice the percentage of any physical neglect (11.35%) as those with college-educated parents (5.68%). Children of parents with college education (or higher) also had significantly lower percentages of any supervisory neglect than all the lower education groups. No significant associations were evident with respect to education and other forms of maltreatment (i.e., physical and sexual abuse).
Household income was not significantly related to either form of neglect or other forms of maltreatment. However, perceived financial strain reported by parents/caregivers was significantly associated with physical neglect. Pairwise comparisons showed that children of parents who responded “very true” to the statement: “You don’t have enough money for clothes and household items that you or your family need” had significantly higher percentages any physical neglect than did children of parents who responded “not true” or “a little true.” Perceived financial strain was not significantly related to supervisory neglect or to other forms of maltreatment. Families who were currently receiving some form of public assistance also had higher percentages of physical neglect, but, again, this indicator socioeconomic of resources was unrelated to supervisory neglect and other forms of maltreatment.
The socioeconomic resource/deficit indicator that appeared to be most strongly related to both child neglect and other forms of maltreatment is a history of parental job loss. Children and youth whose parents who have been unemployed were substantially more likely to be exposed to any physical neglect (13.54%) and any supervisory neglect (19.08%) than those without histories of employment problems (5.19% and 8.58%, respectively). This indicator was also significantly related to other forms of maltreatment.
Child Neglect and Other Victimization Risk
Table 3 presents relative risk ratios (RRR) for the likelihood of exposure to (a) different forms of maltreatment (physical, emotional, sexual) and (b) eight other forms of (nonmaltreatment) victimization, given exposure to a particular category of neglect. With respect to maltreatment, both any physical neglect and any supervisory neglect were related to substantially greater risk of physical abuse, emotional abuse, and sexual abuse. Of particular note was the strong relationship between physical neglect and sexual abuse, with any physical neglect associated with 9.07 times the risk of sexual abuse, relative to children who never experienced physical neglect. Any physical neglect also increased the relative risk of physical abuse by 5.28 times and emotional abuse by 3.5 times. Similarly, any supervisory neglect was most strongly related to sexual abuse, associated with a 4.7-fold increase in the risk of that form of maltreatment, relative to children with no supervisory neglect exposure. Any supervisory neglect also increased the relative risk of physical abuse by 3.76 times and emotional abuse by 3.4 times.
With respect of other types of victimization, both forms of neglect were significantly related to virtually every form of victimization that was assessed. The strongest associations with any physical neglect was witnessing sibling abuse (RRR = 9.6). Any supervisory neglect was most strongly related to sexual victimization by an adult noncaregiver, increasing the risk of this form of victimization by 7.51 times, relative to children with no supervisory neglect.
Child Neglect and Trauma Symptoms
Table 4 presents multivariate analyses predicting trauma symptoms. Models 1 and 2 show the effects of any supervisory neglect and any physical neglect (separately) on current trauma symptoms, controlling for sociodemographic factors (child age, gender, SES, and race). The size of the standardized regression coefficients is similar (βs = .27 and .23, respectively). Model 3 shows the effect of each category of neglect, independent of the other, and controlling for exposure to any other form of maltreatment and exposure to any other type of victimization (see Table 3 for types included). Physical neglect, supervisory neglect, any other maltreatment, and any (nonmaltreatment) victimization exposure, all have independent effects on trauma symptoms (p < .001), controlling for demographic factors. In Model 4, any other maltreatment and any other victimization exposure were dropped from the equation and poly-victimization (total number of victimization types experienced, not including neglect types) were added. Poly-victimization had a very strong association with trauma symptoms (B = 0.62; p < .001), and when poly-victimization was controlled, the coefficients for supervisory and physical neglect were no longer statistically significant.
Discussion
Child neglect has been the least studied form of maltreatment. Moreover, the large majority of research addressing this form of victimization has relied on data from CPS and, as such, inevitably captures only a small fraction of child neglect experiences among U.S. children and youth. In contrast, the current study assessed different forms of neglect in a large nationally representative sample, allowing the examination of how neglect is linked to components of socioeconomic resources in the general population and how it may be associated with risk of other victimization both within and outside of the family context.
Unlike CPS statistics that generally show much higher rates of neglect among younger children (U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, & Children’s Bureau, 2015), we found relatively similar rates of physical neglect among younger and older kids and even higher rates of supervisory neglect among older youth. This suggest that, while more severe cases of neglect among very young children are particularly likely to come to the attention of authorities, our survey measures may better capture exposure to more moderate levels of victimization across a wider developmental spectrum of childhood.
Findings suggest that household income is not a good predictor of child neglect. Parent education, in contrast, predicted both physical neglect and supervisory neglect. Consistent with a substantial literature linking parental education and positive caregiving skills and styles (Bornstein & Bradley, 2003; Davis-Kean, 2005; Harding, Morris, & Hughes, 2015), this suggests that neglect may be less linked to income resources than it is to potential deficits in parenting knowledge and abilities. The particular significance of parent education for child neglect is further highlighted by the lack of association between parent education and other forms of maltreatment. Thus, parent’s understanding of children’s developmental needs and possible hazards in the home may be especially relevant for preventing neglect.
Indicators of poverty and economic adversity were strongly associated with physical neglect but unrelated to supervisory neglect and other forms of maltreatment. Children whose parents responded “very true” to the statement, “there is not enough money for clothes and household items that you or your family need,” had about 2½ times the rate of physical neglect as did children whose parents responded “not true” to the statement. A similar pattern was evident for families who reported currently receiving some form of public assistance (e.g., TANF, Medicaid, food stamps). The associations between these factors and physical neglect (but not supervisory neglect or other maltreatment) point to a direct impact of economic deprivation in reducing parent’s ability to provide adequate food, clothing, shelter, and hygiene-related care to children. In contrast, economic deprivation in itself appears less problematic for adequate supervision and does not increase risk of physical or sexual abuse in this nationally representative sample. The lack of association between household income and physical neglect, however, is interesting and suggests that either income is a poor indicator of economic deprivation in our sample or that it is exposure to economic-related stressors (e.g., perceived financial strain and reliance on public assistance) that is the most influential aspect of poverty. History of parental unemployment was the only socioeconomic indicator that was significantly associated with all forms of maltreatment, suggesting a possible link between the stress of job loss and maltreatment. Moreover, given research demonstrating strong linkages between unemployment and problematic substance and alcohol use (Henkel, 2011), history of parental unemployment may represent a more general indicator of parental dysfunction that contributes to risk for all forms of maltreatment (Walsh, MacMillan, & Jamieson, 2003).
It is exceedingly clear from this study that exposure to physical and supervisory neglect is strongly linked to increased risk for other types of maltreatment and other forms of nonmaltreatment victimization. Interestingly, physical neglect appeared to be particularly strongly associated with sexual abuse and witnessing sibling abuse, and supervisory neglect was most strongly associated with risk for sexual victimization by a (nonfamily) adult. Although the exact reasons for these particular associations are not entirely evident, the linkages between neglect and risk of sexual victimization by adults (both family and nonfamily) suggest the need for practitioners to pay particular attention to the possibility of these types of serious exposures among neglected youth.
Exposure to physical neglect and exposure to supervisory neglect each has independent and unique effects on trauma symptoms, after accounting for any co-occurring maltreatment (yes/no) and any other forms of victimization (yes/no). Although the size of their relative associations with trauma is smaller than exposure to other individual forms of victimization, these findings demonstrate that both categories of neglect are consequential for mental health even when other individual victimization types are also experienced. That being said, findings also showed that poly-victimization, the total number of different victimizations to which children and youth are exposed, supersedes the individual effects of neglect. That is, individual effects of neglect are no longer evident when we consider for the full cumulative burden of different victimizations experienced by neglected children. Indeed, poly-victimization showed a very strong association with trauma symptom levels, with neglect and sociodemographic factors controlled. This is consistent with previous research demonstrating that poly-victimization substantially accounts for the effects of other individual victimizations types such as sexual assault, bullying, or physical abuse (Finkelhor et al., 2007; Turner, Finkelhor, & Ormrod, 2010). Thus, among neglected children and youth, it is the poly-victims, children, and youth for which neglect represents only one of many different types of victimization exposures, who experience the most trauma symptoms.
Limitations and Future Research
Despite the strength of this study, including the use of a large nationally representative sample and comprehensive assessments of victimization across a wide developmental spectrum, there are also a number of important limitations. First, it is important to remember that in identifying clusters of victimization exposure that involve neglect, we cannot speak to the temporal ordering of those experiences since all victimizations were assessed as lifetime events. The results address how neglect types and other victimizations are associated with greater risk for other types but do not demonstrate that neglect “causes” other victimization exposures.
Second, as others have observed, the assessment of neglect in a population-based survey is particularly challenging as it is more difficult to capture a lack of appropriate behavior than it is to capture the occurrence of a victimization incident. More so than other forms of victimization, it may be difficult to assess exposure to neglect across all ages of childhood and adolescence using a uniforms set of items. Although there is considerable value in using comparable measures to examine variations in exposure across subsamples, the nature of neglect itself is likely to differ across age. We acknowledge, for example, the possibility that our item “parental absence” may include some “false positives” among older youth. Future research would benefit greatly from more methodological studies that directly address measurement issues, including the potential value of integrating different types of severity and/or chronicity thresholds as well as validity across age groups.
Finally, we also acknowledge the possibility that neglect, like other forms of maltreatment, may be underreported in surveys of this type and that age differences in rates of neglect could, to some extent, reflect reporter differences (parent proxy vs. youth) and not only developmental differences.
Conclusions
Our findings suggest that physical neglect is directly linked to economic stressors, while parental education is consequential for both physical and supervisory neglect. These socioeconomic indicators were not similarly important for the risk of other forms of maltreatment. Policies and intervention designed to reduce economic adversity and, importantly, provide parents with a better understanding of children’s needs may provide unique benefits for reducing neglect. Findings also point to the importance of assessing for a broad range of victimizations, both inside and outside the family context, when confronted with cases of child neglect. Indeed, neglect very often represents only one of a whole constellation of victimizations that youth experience. This study also confirms the particular importance to identifying poly-victims among those experiencing neglect. Assessing multiple forms of victimization, not only other forms of maltreatment, would permit the identification of the most at-risk children and encourage treatment that targets the full range of victimizations to which children are exposed. Interventions that focus on reducing neglect, without attention to children’s complete victimization profile, may fail to identify the contexts that place children at greatest risk.
Footnotes
Appendix
Sample Characteristics and Victimization Rates.
| Weighted % | N | Weighted Mean (SE) | Range | |
|---|---|---|---|---|
| Child age (years) | 9.6 (.11) | 2–17 | ||
| 2–9 | 48.61 | 3,581 | ||
| 10–17 | 51.39 | 4,271 | ||
| Gender | ||||
| Male | 51.37 | 4,069 | ||
| Female | 48.63 | 3,783 | ||
| SES | ||||
| Low | 27.61 | 1,610 | ||
| Mid | 57.51 | 4,620 | ||
| High | 14.88 | 1,622 | ||
| Education | ||||
| Less than high school | 6.24 | 384 | ||
| High school or some college | 49.28 | 3,035 | ||
| BA or beyond | 44.48 | 4,432 | ||
| Income | ||||
| Less than US$20,000 | 43.68 | 3,021 | ||
| US$20 to less than 50,000 | 28.01 | 2,320 | ||
| US$50 to less than 100,000 | 15.67 | 1,229 | ||
| US$100,000 or more | 12.64 | 973 | ||
| Financial strain (not enough money for clothes or household items) | ||||
| Very true | 9.94 | 671 | ||
| A little | 18.95 | 1,232 | ||
| Not true | 71.12 | 5,921 | ||
| Currently receiving public assistance (e.g., TANF, Medicaid, food stamps) | ||||
| Yes | 26.16 | 1,663 | ||
| No | 78.74 | 6,160 | ||
| Parent job loss | ||||
| Yes | 25.41 | 1,891 | ||
| No | 74.59 | 5,925 | ||
| Other forms of maltreatment | ||||
| Physical abuse | 9.72 | 683 | ||
| Emotional abuse | 15.28 | 1,189 | ||
| Sexual abuse | 0.56 | 41 | ||
| Other forms of victimization | ||||
| Witness partner violence | 24.59 | 1,666 | ||
| Witnessing sibling abuse/other family violence | 3.58 | 278 | ||
| Peer physical assault | 30.19 | 2,284 | ||
| Peer sexual victimization | 8.02 | 661 | ||
| Assault by an adult (nonrelative) | 1.71 | 137 | ||
| Sexual victimization by an adult (nonrelative) | 0.96 | 73 | ||
| Dating violence | 4.23 | 154 | ||
| Property crime | 40.74 | 3,261 | ||
| Poly-victimizationa | 5.51 (.13) | 0–37 | ||
Note. TANF = Temporary Aid to Needy Families..
aPoly-victimization is measured as a count of the total number of victimization types experienced (neglect items excluded).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by grants from the United States Department of Justice. Office of Justice Programs. Office of Juvenile Justice and Delinquency Prevention (2009-JW-BX-0060 (2009-2011) and (2010-JF-FX-0001 (2010-2013)), and the United States Department of Justice. Office of Justice Programs. National Institute of Justice (2010-IJ-CX-0021 (2011-2013)).