chapter 4
HOW MANY CHILDREN ARE MALTREATED?

Statistics on child abuse are difficult to interpret and compare because there is very little consistency in how information is collected. The definitions of abuse vary from study to study, as do the methods of counting incidents of abuse. Some methods count only reported cases of abuse. Some statistics are based on estimates projected from a small study, while others are based on interviews. In addition, it is virtually impossible to know the extent of child maltreatment that occurs in the privacy of the home.

INCIDENCE AND PREVALENCE OF CHILD MALTREATMENT

Researchers use two terms—incidence and prevalence—to describe the estimates of the number of victims of child abuse and neglect. Andrea J. Sedlak and Diane D. Broadhurst defined incidence as the number of new cases occurring in the population during a given period (Third National Incidence Study of Child Abuse and Neglect [NIS-3], U.S. Department of Health and Human Services, National Center on Child Abuse and Neglect, Washington, DC, 1996). The incidence of child maltreatment is measured in terms of incidence rate: the number of children per one thousand children in the U.S. population who are maltreated annually. Surveys based on official reports by child protective services (CPS) agencies and community professionals are a major source of incidence data.

Prevalence, as defined by NIS-3, refers to the total number of child maltreatment cases in the population at a given time. Some researchers use lifetime prevalence to denote the number of people who have had at least one experience of child maltreatment in their lives. To measure the prevalence of child maltreatment, researchers use self-reported surveys of parents and child victims. Examples of self-reported surveys are the landmark 1975 National Family Violence Survey and the 1985 National Family Violence Resurvey conducted by Murray A. Straus and Richard J. Gelles.

Official Reports

Studies based on official reports depend on a number of things happening before an incident of abuse can be recorded. The victim must be seen by people outside the home, and these people must recognize that the child has been abused. Once they have recognized this fact, they must then decide to report the abuse and find out where to report it. Once CPS receives and screens the report for appropriateness, it can then take action.

In some cases the initial call to CPS is prompted by a problem that needs to be handled by a different agency. It may be a case of neglect due to poverty rather than abuse, although the initial report is still recorded as abuse.

For the data to become publicly available, CPS must keep records of its cases and then pass them on to a national group that collects those statistics. Consequently, final reported statistics are understated estimates—valuable as indicators but not definitive findings. It is very unlikely that accurate statistics on child abuse will ever be available.

COLLECTING CHILD MALTREATMENT DATA

The 1974 Child Abuse Prevention and Treatment Act (CAPTA; Public Law 93-247) created the National Center on Child Abuse and Neglect (NCCAN) to coordinate nationwide efforts to protect children from maltreatment. As part of the former U.S. Department of Health, Education, and Welfare, NCCAN commissioned the American Humane Association (AHA) to collect data from the states. The first time the AHA collected data, in 1976, it recorded an estimated 416,000 reports, affecting 669,000 children. Between 1980 and 1985 the AHA reported a 12% annual increase in maltreatment reports to CPS agencies. By 1990 reports of child maltreatment had risen to 1.7 million, affecting about 2.7 million children.

In 1985 the federal government stopped funding data collection on child maltreatment. In 1986 the National Committee to Prevent Child Abuse (NCPCA; now called Prevent Child Abuse America) picked up where the government left off. The NCPCA started collecting detailed information from the states on the number of children abused, the characteristics of child abuse, the number of child abuse deaths, and changes in the funding and extent of child welfare services.

In 1988 the Child Abuse Prevention, Adoption and Family Services Act (Public Law 100-294) replaced the 1974 CAPTA. The new law mandated that NCCAN, as part of the U.S. Department of Health and Human Services (HHS), establish a national data collection program on child maltreatment. In 1990 the National Child Abuse and Neglect Data System (NCANDS), designed to fulfill this mandate, began collecting and analyzing child maltreatment data from CPS agencies in the fifty states and the District of Columbia. The first three surveys were known as Working Paper 1, Working Paper 2, and Child Maltreatment 1992. NCANDS has since conducted the survey Child Maltreatment. The latest survey was Child Maltreatment 2002, released in April 2004.

As part of the 1974 CAPTA, Congress also mandated NCCAN to conduct a periodic National Incidence Study of Child Abuse and Neglect (NIS). Data on maltreated children are collected not only from CPS agencies but also from professionals in community agencies, such as law enforcement, public health, juvenile probation, mental health, and voluntary social services, as well as from hospitals, schools, and day care centers. The NIS is the single most comprehensive source of information about the incidence of child maltreatment in the United States, because it analyzes the characteristics of child abuse and neglect that are known to community-based professionals, including those characteristics not reported to CPS. The latest study, NIS-3, was released in 1996. In 2003 the Keeping Children and Families Safe Act (Public Law 108-36) directed the collection of data for NIS-4.

Pursuant to the CAPTA Amendments of 1996 (Public Law 104-235), NCCAN ceased operating as a separate agency. Since then all child maltreatment prevention functions have been consolidated within the Children's Bureau of the HHS.

CPS MALTREATMENT REPORTS

Collecting child maltreatment data from the states is difficult because each state has its own method of gathering and classifying the information. Most states collect data on an incident basis; that is, they count each time a child is reported for abuse or neglect. If the same child is reported several times in one year, each incident is counted. Consequently, the number of incidents of child maltreatment may be greater than the number of maltreated children.

In 2002 CPS agencies received an estimated 2,617,000 referrals, or reports, alleging the maltreatment of about 4.5 million children (Child Maltreatment 2002, U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau, Washington, DC, 2004), which may include some children who were reported and counted more than once. States may vary in the rates of child maltreatment reported. States differ not only in definitions of maltreatment but also in the methods of counting reports of abuse. Some states count reports based on the number of incidents or the number of families involved, rather than on the number of children allegedly abused. Other states count all reports to CPS, while others count only investigated reports.

In 2002 forty-two states submitted child-level data for each report of alleged maltreatment. The data include, among other things, the demographics about the children and the perpetrators, types of maltreatment, and dispositions (findings after investigation or assessment of the case). The remaining eight states and the District of Columbia submitted only summary statistics, such as the number of child victims of maltreatment. CPS agencies screened in (accepted for further assessment or investigation) 1,141,820 referrals, or 67.1%, of referrals. Overall, the rate of maltreatment referrals ranged from 12.8 per one thousand children (Pennsylvania) to 78.1 per one thousand children (Montana) under age eighteen. (See Table 4.1.)

Dispositions of Investigated Reports

After a CPS agency screens in a report of child maltreatment, it initiates an investigation. Some states follow one time frame for responding to all reports, while others follow a priority system, investigating high-priority cases within one to twenty-four hours. According to the National Child Abuse and Neglect Data System (NCANDS), in 2002 twenty-three states that reported response time showed an average response time of fifty-two hours.

Following investigation of the report of child maltreatment, the CPS agency assigns a disposition, or finding, to the report. Prior to 2000, reports of alleged child maltreatment received one of three dispositions—indicated, substantiated, or unsubstantiated. In 2000 several states announced plans to establish an alternative response program to reports of alleged child maltreatment. If the child is at a serious and immediate risk of maltreatment, CPS responds with the traditional formal investigation, which may involve removing the child from the home. If it is determined, however, that the parent likely will not endanger the child, CPS workers use the alternative response to help the family. This involves a more informal approach. Instead of removing the child from the home environment, CPS steps in to assist the whole family by, for example, helping reduce stress that may lead to child abuse through provision of child care, adequate housing,

TABLE 4.1

and education in parenting skills. In 2002 ten states implemented the alternative response program. NCANDS used the following dispositions for its 2002 report:

• A disposition of "substantiated" means that sufficient evidence existed to support the allegation of maltreatment or risk of maltreatment.
• A disposition of "indicated or reason to suspect" means that the abuse and/or neglect could not be confirmed, but there was reason to suspect that the child was maltreated or was at risk of maltreatment.
• A disposition of "unsubstantiated" means that no maltreatment occurred or sufficient evidence did not exist to conclude that the child was maltreated or was at risk of being maltreated.
• A disposition of "alternative response victim" means that, when a response other than investigation was provided, the child was identified as a victim of maltreatment.
• A disposition of "alternative response nonvictim" means that, when a response other than investigation was provided, the child was not identified as a victim of maltreatment.

Of the more than 1.1 million reports that were investigated, 60.4% were unsubstantiated. More than one-fourth (26.8%) were substantiated, and 3.5% were indicated. About 0.1% were determined alternative response victim, and 4.7% were alternative response nonvictim. Dispositions that were identified "closed with no finding" referred to cases in which the investigation could not be completed because the family moved out of the jurisdiction, the family could not be found, or the needed reports were not filed within the required time limit. Such dispositions accounted for 1.7%. (See Figure 4.1.)

VICTIMS OF MALTREATMENT

Rates of Victimization

In 2002 an estimated 896,000 children were victims of maltreatment in the United States, down from 903,000 in 2001. A total of 12.3 children for every one thousand children in the population were victims of abuse or neglect. While this rate of victimization was higher than that for 1999 (11.8 per one thousand children), it was lower than the rates for the previous years in the 1990s. The rate of maltreatment had peaked at 15.3 per one thousand children in 1993. (See Table 4.2.)

Types of Maltreatment

According to Child Maltreatment 2002 more than three out of five maltreated children (60.5%) suffered neglect (including medical neglect). Nearly one-fifth (18.6%) were physically abused, and 9.9% were sexually abused. An additional 6.5% were subjected to emotional, or psychological, maltreatment. Another 18.9% experienced other types of maltreatment, including abandonment, congenital drug addiction, and threats to harm a child. Some children were victims of more than one type of maltreatment.

rates of maltreatment by type. Between 1998 and 2002 the rates of the different types of maltreatment

FIGURE 4.1

fluctuated from year to year. The rate of neglect dropped from 6.9 per one thousand children in the population in 1998 to 6.5 per one thousand children in 1999, rising to 7.3 per one thousand children in 2000. The rate was about the same (7.2 per one thousand children) in 2002. The rates for medical neglect remained constant during the five-year period. The rates for physical abuse and sexual abuse declined, from 2.9 to 2.3 per one thousand children for the former, and from 1.5 to 1.2 per one thousand children for the latter. The rate for psychological abuse remained the same for 1998 and 2002 (0.8 per one thousand children). (See Figure 4.2.)

Gender and Age of Victims

In 2002 more female children (51.9%) were maltreated than their male counterparts (48.1%). Younger children represented most of the maltreated victims. Generally, the older the child gets, the more the likelihood of abuse decreases. The victimization rate for infants and toddlers up through age three was 16 per one thousand children of the same age group, compared to 13.7 per one thousand for children ages four to seven. The rate of victimization for children ages eight to eleven was 11.9 per one thousand; for ages twelve to fifteen, 10.6 per one thousand; and for ages sixteen and seventeen, six per one thousand. (See Figure 4.3.)

Race and Ethnicity of Victims

In 2002 more than half (54.2%) of maltreatment victims were white, and one-quarter (26.1%) were African-American. Hispanics, who may be of any race, accounted for 11% of the victims. Native Americans (1.8%) and

TABLE 4.2

FIGURE 4.2

Asians/Pacific Islanders (0.9%) made up the lowest proportions of victims. (See Figure 4.4.)

PERPETRATORS OF CHILD MALTREATMENT

The law considers perpetrators to be those persons who abuse or neglect children under their care. They may be parents, foster parents, other relatives, or other caretakers. In 2002 more than two of five victims (40.3%) were maltreated by their mother acting alone. Another 19.1% experienced maltreatment from their father acting alone. Eighteen percent were maltreated by both parents. About 5.4% of the victims were maltreated by their mother and another person whose relationship with the mother was not known. One percent was maltreated by their father and another person

FIGURE 4.3

whose relationship to the father was not known. Thirteen percent were maltreated by nonparental perpetrators, while 3.2% were maltreated by unknown individuals.

More than three-quarters (77.9%) of perpetrators were under age forty, with one-third between the ages of twenty and twenty-nine. Perpetrators were more likely to be female (58.3%) than male (41.7%).

Parents (including birth parents, adoptive parents, and stepparents) were responsible for the most cases of neglect, followed by child daycare providers and foster parents. Other relatives perpetrated more than a quarter of sexual abuse cases, while residential facility staff and unmarried partners of parents were the most likely to inflict physical abuse on children. (See Figure 4.5.)

DEATHS FROM CHILD MALTREATMENT

Child fatality is the most severe result of abuse and neglect. In 2002 CPS and other state agencies, including coroners' offices and fatality review boards, reported an estimated 1,390 deaths from child maltreatment, up from 1,373 deaths in 2001 and 1,306 deaths in 2000. The 2002 national fatality rate was nearly two deaths per one hundred thousand children in the general population. The District of Columbia reported the highest rate (11.6 per one hundred thousand), followed by West Virginia (7.5 per one hundred thousand) and Missouri (3.8 per one hundred thousand). (See Table 4.3.)

Children younger than four years of age accounted for a majority (76.1%) of deaths. Of these, infants younger than a year old comprised 41.2% of the fatalities.

FIGURE 4.4

More males under age one (18.8 per one hundred thousand boys of the same age) died from maltreatment than did females under age one (12.4 per one hundred thousand girls of the same age). Very young children are more likely to be victims of child fatalities because of their small size, their dependency on their caregivers, and their inability to defend themselves.

Neglect alone was responsible for more than one-third (37.6%) of maltreatment deaths. More than onequarter (29.9%) of fatalities resulted from physical abuse. Another 28.9% of fatalities resulted from a combination of maltreatment types. (See Figure 4.6.) States also provided data on the victims' prior contact with CPS agencies. About 12% of the victims' families had received family preservation services during the five years before the deaths occurred.

Perpetrators of Fatalities

The 2002 report showed nearly four of five maltreatment deaths (78.9%) were inflicted by one or both parents of the victims. Mothers alone accounted for about one-third (32.6%) of the deaths, while fathers were the perpetrators in 16.6% of the deaths. In about one-fifth (19.2%) of cases, both parents were responsible for causing their children's death.

According to the National Clearinghouse on Child Abuse and Neglect Information (Child Abuse and Neglect Fatalities: Statistics and Interventions, Washington, DC, 2004), persons who commit fatal child abuse are often younger adults in their mid-twenties who did not finish high school and are living at or below the poverty level. They are likely to suffer from depression and are unable to cope with stress.

FIGURE 4.5

FIGURE 4.6

Child Fatality Review Teams

Historically, law enforcement, child protection agencies, and public health agencies worked separately in investigating child maltreatment deaths. In response to the increasing number of child deaths, all states have created multidisciplinary child fatality review teams to investigate the deaths and develop solutions to support families in crisis. These teams consist of prosecutors, medical examiners, law enforcement personnel, CPS personnel, health care providers, and other professionals.

State teams are formed to work with local teams, which are responsible for the management of individual cases. The number of child deaths reviewed by a local team depends on the county size. A local team in a large county may review just cases referred by the medical examiner, while a team in a smaller county may review child deaths from all causes.

TABLE 4.3

Are Child Maltreatment Fatalities Properly Reflected in Death Certificates?

Although child fatality review teams have been formed in every state and the District of Columbia, NCANDS remains the sole national system that tracks child maltreatment deaths, typically just those cases that reach CPS. Experts believe that there are likely more deaths each year due to child abuse and neglect than are reported to CPS and other agencies. According to researchers Tessa L. Crume, Carolyn DiGuiseppi, Tim Byers, Andrew P. Sirotnak, and Carol J. Garrett, although the federal government had concluded in 1993 that death certificates underreported child maltreatment fatalities, to date it has not done anything to remedy the problem. To determine whether child maltreatment is ascertained in death certificates, the researchers compared data collected by a child fatality review committee (CFRC) on child fatalities in Colorado between 1990 and 1998 with the death certificates issued for those fatalities ("Underascertainment of Child Maltreatment Fatalities by Death Certificates, 1990–1998," Pediatrics, vol. 110, no. 2, August 2002).

Crume and her colleagues found that only half of the maltreatment deaths were ascertained by death certificates. Of the 295 deaths confirmed by the CFRC to have resulted from maltreatment, just 147 were noted in the death certificates as such. Female children and non-Hispanic African-American children were more likely to be linked to higher ascertainment in the death certificates. Maltreatment was also more likely to be confirmed as the contributing factor when the death involved violence, such as bodily force, the use of firearms, or the use of sharp or blunt objects. A lower proportion of deaths (less than 20%) was attributed to less obvious child maltreatment, including neglect and abandonment.

NATIONAL INCIDENCE STUDY OF CHILD ABUSE AND NEGLECT

The National Incidence Study of Child Abuse and Neglect (NIS) was a congressionally mandated periodic survey of child maltreatment. The results of the first NIS were published in 1981, and those of the second NIS (NIS-2) in 1988. The most recent NIS, the Third National Incidence Study of Child Abuse and Neglect, (NIS-3) (U.S. Department of Health and Human Services, National Center on Child Abuse and Neglect, Washington, DC), was released in 1996. NIS-3 differed from the annual Child Maltreatment reports because NIS-3 findings were based on a nationally representative sample of more than 5,600 professionals in 842 agencies serving forty-two counties. NIS-3 included not only child victims investigated by CPS agencies, but also children seen by community institutions (such as daycare centers, schools, and hospitals) and other investigating agencies (such as public health departments, police, and courts). In addition, victim counts were unduplicated, which means that each child was counted only once.

Definition Standards

NIS-3 used two standardized definitions of abuse and neglect:

• Harm Standard—required that an act or omission must have resulted in demonstrable harm in order to be considered as abuse or neglect
• Endangerment Standard—allowed children who had not yet been harmed by maltreatment to be counted in the estimates of maltreated children if a non-CPS professional considered them to be at risk of harm or if their maltreatment was substantiated or indicated in a CPS investigation

Incidence of Maltreatment

In 1993, under the Harm Standard, an estimated 1,553,800 children were victims of maltreatment, a 67% increase from the NIS-2 estimate (931,000 children) and a 149% increase from the first NIS estimate (625,100 children). Significant increases occurred for all types of abuse and neglect, as compared with the two earlier NIS surveys. The more than 1.5 million child victims of maltreatment in 1993 reflected a yearly incidence rate of 23.1 per one thousand children under age eighteen, or about one in forty-three children. (See Table 4.4.)

In 1993, under the Endangerment Standard, an estimated 2,815,600 children experienced some type of maltreatment. This figure nearly doubled the NIS-2 estimate of 1,424,400. As with the Harm Standard, marked increases occurred for all types of abuse and neglect. The incidence rate was 41.9 per one thousand children under age eighteen, or approximately one in twenty-four children. (See Table 4.5.)

comparison of maltreatment estimates under the two standards. In 1993 the Endangerment Standard included an additional 1,261,800 children under age eighteen (an 81% difference) beyond those counted under the stricter Harm Standard. This means that children included under the Harm Standard represented 55% of those counted under the Endangerment Standard. Harm Standard children accounted for 61% of the Endangerment Standard total of all abused children, and 45% of the Endangerment Standard total of all neglected children.

Characteristics of Abused Children

gender. Under both the Harm and Endangerment Standards, more females were subjected to maltreatment than males. Females were sexually abused about three times more often than males. Males, however, were more likely to experience physical and emotional neglect under the Endangerment Standard. Under both standards, males suffered more physical and emotional neglect, while females suffered more educational neglect. Males were at a somewhat greater risk of serious injury and death than females. (See Table 4.6 and Table 4.7.)

age. NIS-3 found a lower incidence of maltreatment among younger children, particularly ages zero to five. This may be due to the fact that, prior to reaching school age, children are less observable to community professionals, especially educators—the group most likely to report suspected maltreatment. In addition, NIS-3 noted a disproportionate increase in the incidence of maltreatment among children between the ages of six and fourteen. (See

TABLE 4.4

TABLE 4.5

Figure 4.7 and Figure 4.8.) Sedlak and Broadhurst, the authors of NIS-3, noted a lower incidence of maltreatment among children older than fourteen years. Older children

TABLE 4.6

are more likely to escape if the abuse becomes more prevalent or severe. They are also more able to defend themselves and/or fight back.

TABLE 4.7

Under the Harm Standard, only ten per one thousand children in the zero to two age group experienced overall maltreatment. The numbers were significantly higher for children ages six to seventeen. Under the Endangerment Standard, twenty-six per one thousand children ages zero to two were subjected to overall maltreatment. A slightly higher number of children (29.7 per one thousand children) in the oldest age group (fifteen to seventeen years old) suffered maltreatment of some type. As with the Harm Standard, children between the ages of six and fourteen had a higher incidence of maltreatment. (See Figure 4.8.)

race. NIS-3 found no significant differences in race in the incidence of maltreatment. The authors noted that this finding may be somewhat surprising, considering the over-representation of African-American children in the child welfare population and in those served by public agencies. They attributed this lack of race-related difference in maltreatment incidence to the broader range of children identified by NIS-3, compared with the smaller number investigated by public agencies and the even smaller number receiving child protective and other welfare services. NIS-2 also had not found any disproportionate differences in race in relation to maltreatment incidence.

Family Characteristics

family structure. Under the Harm Standard, among children living with single parents, an estimated 27.3 per one thousand under age eighteen suffered some

FIGURE 4.7

FIGURE 4.8

type of maltreatment—almost twice the incidence rate for children living with both parents (15.5 per one thousand). The same rate held true for all types of abuse and neglect. Children living with single parents also had a greater risk of suffering serious injury (10.5 per one thousand) than

TABLE 4.8

did those living with both parents (5.8 per one thousand). (See Table 4.8.)

Under the Endangerment Standard, an estimated 52 per one thousand children living with single parents suffered some type of maltreatment, compared with 26.9 per one thousand living with both parents. Children in single-parent households were abused at a 45% higher rate than those in two-parent households (19.6 versus 13.5 per one thousand) and suffered more than twice as much neglect (38.9 versus 17.6 per one thousand). Children living with single parents (10.5 per one thousand) were also more likely to suffer serious injuries than those living with both parents (5.9 per one thousand). (See Table 4.9.)

family size. The number of children in the family was related to the incidence of maltreatment. Additional children meant additional tasks and responsibilities for the parents; therefore, it followed that the rates of child maltreatment were higher in these families. Among children in families with four or more children, an estimated 34.5 per one thousand under the Harm Standard and 68.1 per one thousand under the Endangerment Standard suffered some type of maltreatment.

Surprisingly, households with only one child had a higher maltreatment incidence rate than did households with two to three children (22 versus 17.7 per one thousand children under the Harm Standard, and 34.2 versus 34.1 per one thousand children under the Endangerment Standard). The authors thought that an only child might have been in a situation where parental expectations were all focused on that one child. Another explanation was that "only" children might have been in households where the parents were just starting a family and were relatively young and inexperienced.

family income. Family income was significantly related to the incidence rates of child maltreatment. Under the Harm Standard, children in families with annual incomes less than $15,000 had the highest rate of maltreatment (47 per one thousand). The figure is almost twice as high (95.9 per one thousand) using the Endangerment Standard. Children in families earning less than $15,000 annually also sustained more serious injuries. (See Table 4.10 and Table 4.11.)

Characteristics of Perpetrators

relationship to the child. Most child victims (78%) were maltreated by their birth parents. Parents accounted for 72% of physical abuse and 81% of emotional abuse. Almost half (46%) of sexually abused children, however, were violated by someone other than a parent or

TABLE 4.9

parent-substitute. More than a quarter (29%) were sexually abused by a birth parent, and 25% were sexually abused by a parent-substitute, such as a stepparent or a mother's boyfriend. In addition, sexually abused children were more likely to sustain fatal or serious injuries or impairments when birth parents were the perpetrators. (See Table 4.12.)

perpetrators' genders. Overall, children were somewhat more likely to be maltreated by female perpetrators (65%) than by males (54%). Among children maltreated by their natural parents, most (75%) were maltreated by their mothers, and almost half (46%) were maltreated by their fathers. (Children who were maltreated by both parents were included in both "male" and "female" perpetrator counts.) Children who were maltreated by other parents and parent-substitutes were more likely to have been maltreated by a male (85%) than by a female (41%). Four of five children (80%) who were maltreated by other adults were maltreated by males, and only 14% were maltreated by other adults who were females. (See Table 4.13.) Note that the numbers will not add to 100% because many children were maltreated by both parents.)

Neglected children differed from abused children with regard to the gender of the perpetrators. Because mothers or other females tend to be the primary caretakers, children were more likely to suffer all forms of neglect by female perpetrators (87% versus 43% by male perpetrators). In contrast, children were more often abused by males (67%) than by females (40%). (See Table 4.13.)

SURVEY REVEALS YOUTH'S KNOWLEDGE OF PEER ABUSE

The most recent government survey revealed that, in 2002, an estimated 896,000 children under age eighteen were victims of maltreatment. Steve Crabtree reported that a 2003 nationwide survey of teenagers, ages thirteen to seventeen, found that more than one-third (36%) indicated they knew of a person their age who had been physically or sexually abused ("One-Third of Teens Know of Abuse Among Peers," Gallup Youth Survey, The Gallup Organization, Princeton, NJ, May 20, 2003). Nearly twice as many girls (47%) as boys (25%) said that they knew of a friend or a classmate who had been maltreated. (See Figure 4.9.)

According to Crabtree, the Gallup Youth Survey did not ask teens the direct question of whether or not they themselves had been victims of maltreatment, partly

TABLE 4.10

because young people may be reluctant to reveal their own abuse. Some blame themselves for the abuse. Others feel ashamed of their experience, while still others are afraid of retaliation from the perpetrator. Interestingly, compared to adolescents who did not know of a friend or classmate who had been maltreated, those who revealed they were aware of a peer's experience with abuse were more likely to use alcohol (40% versus 30% of teens with no awareness of peer abuse), to have smoked in the past week (13% versus 6%), and to have tried marijuana (30% versus 15%). (See Table 4.14.)

TABLE 4.11

FIGURE 4.9

TABLE 4.12

TABLE 4.13

TABLE 4.14