INVESTIGACIÓN ORIGINAL ORIGINAL RESEARCH
Dimensiones del castigo infantil en dos países de América Central: Guatemala y El Salvador
Ilene S. SpeizerI, *; Mary M. GoodwinII; Ghazaleh SamandariI; Shin Y. KimII; Maureen ClydeII
IUniversity of North Carolina at Chapel Hill, School of Public Health, Department of Maternal and Child Health, Chapel Hill, NC 27516, United States of America
IICenters for Disease Control and Prevention, Division of Reproductive Health, Atlanta, GA 30341, United States of America
OBJECTIVE: Severe physical punishment of children is an important issue in international child health and welfare. This study examines such punishment in Guatemala and El Salvador.
METHODS: Data came from nationally representative surveys of women aged 1549 and men aged 1559 residing in Guatemala (2002) and El Salvador (20022003). The surveys included questions about punishment experienced during childhood, with response options ranging from verbal scolding to beating. In Guatemala, parents were asked how they disciplined their children; questions allowed them to compare how they were punished in their childhood with how they punished their own children. Bivariate and multivariate analyses are presented.
RESULTS: In Guatemala, 35% of women and 46% of men reported being beaten as punishment in childhood; in El Salvador, the figures were 42% and 62%, respectively. In both countries, older participants were relatively more likely than younger participants to have been beaten as children. Witnessing familial violence was associated with an increased risk of being beaten in childhood. In Guatemala, having experienced physical punishment as a child increased the chance that parents would use physical punishment on their own children. Multivariate analyses revealed that women who were beaten in childhood were significantly more likely in both countries to be in a violent relationship.
CONCLUSIONS: The use of beating to physically punish children is a common problem in Guatemala and El Salvador, with generational and intergenerational effects. Its negative and lingering effects necessitate the introduction of policies and programs to decrease this behavior.
Key words: Child abuse, parenting, domestic violence, Guatemala, El Salvador.
OBJETIVOS: El castigo físico severo de niños es un tema importante de la salud y el bienestar infantil en el mundo. En el presente estudio se analiza este tipo de castigo en Guatemala y El Salvador.
MÉTODOS: Se tomaron los datos de encuestas representativas nacionales realizadas a mujeres de 14 a 59 años y hombres de 15 a 59 años que residían en Guatemala (2002) y El Salvador (20022003). Las encuestas contenían preguntas sobre los castigos que sufrieron en su niñez, con posibilidades de respuesta que iban desde regaños verbales hasta golpizas. En Guatemala se preguntó a los padres cómo disciplinaban a sus hijos; las preguntas les permitían comparar cómo ellos eran castigados en su niñez y cómo ellos castigaban a sus hijos. Se presentan los resultados de los análisis bifactorial y multifactorial.
RESULTADOS: En Guatemala, 35% de las mujeres y 46% de los hombres informaron haber sido golpeados como castigo durante su niñez; en El Salvador, las cifras fueron 42% y 62%, respectivamente. En ambos países, los encuestados más viejos tenían una mayor probabilidad de haber sido golpeados en su niñez que los encuestados más jóvenes. Haber presenciado violencia familiar se asoció con un incremento en el riesgo de recibir golpizas durante la niñez. En Guatemala, haber recibido castigos físicos durante la niñez aumentó la probabilidad de que como padres aplicaran castigos físicos a sus hijos. El análisis multifactorial reveló que las mujeres de ambos países golpeadas durante su niñez tenían significativamente más probabilidades de encontrarse en una relación violenta.
CONCLUSIONES: El uso de golpes para castigar físicamente a niños es un problema frecuente en Guatemala y El Salvador, con efectos generacionales e intergeneracionales. Sus efectos prolongados y negativos requieren de la aplicación de políticas y programas que permitan reducir este comportamiento.
Palabras clave: Maltrato a los niños, patria potestad, violencia doméstica, Guatemala, El Salvador.
Severe physical punishment of children is an important global child health and welfare issue (1). The World Health Organization has defined physical abuse of a child as "those acts of commission by a caregiver that cause actual physical harm or have the potential for harm" (2). Physical punishment refers to the use of physical acts to discipline or punish that can range from spanking, slapping, and shoving to throwing items at the child and kicking or beating the child (3, 4). Views on the acceptability of physical punishment of children vary across cultures. Sweden was the first nation to have instituted a statutory ban on parental use of physical punishment, and approximately 10 countries have followed suit (5). In the United States, efforts to regulate corporal punishment have focused primarily on its use in schools rather than in the family setting (6). In a recent survey conducted by the International Society for the Prevention of Child Abuse and Neglect (ISPCAN) that obtained responses from 64 countries around the globe, public officials were asked about their country's legislation, policies, and programmatic priorities related to child abuse. Physical punishment of children was the behavior least frequently cited by responding officials as being considered child abuse in their country (7, 8). The ISPCAN report suggests that this lack of concern about physical discipline reflects a continued widespread use of physical punishment around the world and its consideration by many cultures and individuals as an appropriate way to discipline children (7). In the Latin American region, which is the geographic focus of this study, few countries have developed policy or programmatic interventions to reduce the physical punishment of children (1).
In part because of a lack of standard research methods, existing research on the prevalence and characteristics associated with the use of physical punishment is limited. In an effort to collect comparable data across countries, the World Studies of Abuse in the Family Environment (WorldSAFE) project conducted standard interviews with samples of mothers from four countries: Chile, Egypt, India, and the Philippines. The WorldSAFE surveys explored mothers' use of harsh and moderate forms of child physical punishment (9). The surveys found that physical punishment of children is common in all four countries, with parental reports of having hit a child with an object ranging from 18% in Chile to 51% in the Philippines (2).
Several studies in Latin American countries confirm that, as in other parts of the world, physical punishment of children is common in the region. In a study of 497 university students in Costa Rica, approximately 80% reported ever having been punished with spanking, more than a quarter reported having been whipped or flogged, and almost 20% said they had been hit with an object (10). Results from the ACTIVA Project (Multicenter Study on Cultural Norms and Attitudes in Selected Cities of Latin America and Spain), which used cross-sectional samples from eight Latin American cities, found that the prevalence of adult use of physical punishment on a child ranged from 20% to 40% (1). In another study, data from parents attending six clinic sites in Chile and Costa Rica revealed that somewhat higher proportions of parents had used physical punishment on a child, with about half of Chilean parents reporting that they sometimes used physical punishment on their children and a fifth of Costa Rican parents stating that they often hit their children to punish them (11).
In the ACTIVA Project, parental characteristics associated with having used physical punishment included young parental age, being female, limited parental education, low socio-economic status, currently caring for young children, believing that corporal punishment is necessary, and having low self-efficacy for alternatives to physical violence (1). Although the ACTIVA study and others have shown that female parents are more likely than male parents to report having physically punished children, these results often do not control for time spent caring for children (women have greater exposure to disciplining children because they are more often at home caring for them).
The ACTIVA results, as well as those from other studies, document that physical punishment of children has negative consequences for children as they age, including aggressive behavior later in childhood, childhood depression, adolescent and adult violence, alcohol and other substance use, and risky sexual behaviors (1, 4, 8, 12). Furthermore, some researchers theorize that young children who face harsh physical punishment from their parents learn to respond with violence to behavior that they dislike in others (13). While this may be true for some people, other research has indicated that many abused children mature into nurturing and responsible adults (3).
Another cause for concern is the evidence suggesting that violence against children is often passed from one generation to the next, with those abused as children becoming abusive parents in their own adulthood. Information from parents in Chile and Costa Rica shows that those who were punished physically while growing up had a greater tendency to perceive physical punishment as effective than those who did not experience such punishment (12). In several studies, a history of childhood abuse was determined to predict parents' physical punishment of their own children (3, 1416). Parental experience of childhood abuse was even associated with a subsequent belief that physical discipline is a parental privilege or right (17). Associations between maltreatment as a child and experiencing violence at the hands of an intimate partner in adulthood have also been established in numerous studies (18, 19), suggesting that the experience of childhood abuse can render a person more prone to being victimized as an adult (20).
The purpose of this study is to explore dimensions of physical punishment in childhood in two Central American countriesGuatemala and El Salvadorusing population-based, nationally representative data for women aged 1549 and men aged 1559. To date, no nationally representative data on the topic for these two countries have been published. This study focuses on four specific points: (1) types of punishment Guatemalan and Salvadoran women and men report having experienced as children, including the proportion who reported severe physical punishment; (2) types of punishment parents have used with their own children (Guatemala only); (3) intergenerational effects of physical punishment (Guatemala only); and (4) for women, whether severe physical punishment experienced in childhood is associated with being in a violent relationship as an adult (both countries).
Understanding the dimensions and consequences of punishment during childhood in these two Central American countries constitutes a critical first step in raising awareness about the frequency of severe forms of this activity. This information can provide a basis for forming programs and policies to protect children from abusive punishment and to support adults in developing and practicing healthier rearing of children.
MATERIALS AND METHODS
Our study examines population-based, nationally representative data from surveys in two Central American countries: the El Salvador National Family Health Survey, 20022003 (FESAL) (21), and the Guatemala National Maternal and Child Health Survey, 2002 (ENSMI) (22). Both surveys were funded primarily by the United States Agency for International Development, with technical assistance provided by the Centers for Disease Control and Prevention. In Guatemala, the survey was implemented by the University del Valle of Guatemala; in El Salvador, it was implemented by the Demographic Association of El Salvador. Both surveys used multistage sampling designs in which the primary sampling units (PSUs) were census segments. The probability of selection of a PSU was proportional to its size. In each PSU, 30 households were randomly selected. Female and male samples were independent; only one eligible woman or one eligible man was interviewed in each selected household. Eligibility was defined solely by age: 1549 years for women and 1559 years for men. For both surveys, participants were interviewed in or around their homes. Women were interviewed by women and men by men. Before beginning the survey, interviewers requested verbal consent to complete the interview. All participants were assured they did not have to participate and could refuse to answer any of the questions. Response rates were high: in Guatemala and El Salvador, respectively, 94% and 91% of identified eligible women participated, as did 80% and 81% of eligible men. In Guatemala, 9 155 women and 2 538 men participated; in El Salvador, these figures were 10 689 and 1 315, respectively. Differences in the sample sizes for men and women were due to different stratification schemes and the numbers needed for accuracy of indicatorsfor example, the female surveys were designed to obtain regional-level data, whereas male surveys obtained data stratified by area of residence only (urban, suburban, rural); also, more women were needed to obtain accurate estimates of certain indicators such as infant mortality. The data were weighted by the number of eligible respondents in the household.
The surveys cove red a variety of topics on maternal and child health, including pregnancy and childbearing; family planning; access to reproductive health care services; sexual behavior, knowledge, and attitudes; infant mortality and morbidity; breastfeeding and child nutrition; immunizations; attitudes and knowledge about HIV/AIDS; and family violence. In both countries, participants were asked questions about violence only if there was no one older than 2 years present during the interview, a rule meant to provide a greater sense of privacy, confidentiality, and safety for participants. In Guatemala, 97% of women (N = 8 860) and the same percentage of men (N = 2 459) were asked the questions on violence. For El Salvador, the comparable percentages were 88% (N = 9 430) and 95% (N = 1 255).
Measurement of childhood punishment
Questions on childhood punishment were asked as part of the module on family violence. In El Salvador, participants were asked: "Before you were 18 years old, how were you usually punished?" Precoded response options included never punished; beaten with a belt, stick, or cord; reprimanded verbally; prohibited something; locked up; and burned. In Guatemala, participants were asked: "How were you usually punished by your parents?" Precoded responses differed slightly from those in El Salvador and included spanked, reprimanded verbally, prohibited from having or doing something, beaten, locked up, burned, and never punished. Our analysis focused on severe physical punishment, which we defined as having been "beaten with a belt, stick, or cord" for El Salvador and "beaten" for Guatemala. The prevalence of other severe forms of punishment such as burning is presented. However, because these responses were rarely reported, they are not included in the more indepth analyses of specific forms of punishment. Guatemalan participants who had children were asked whether they thought physical punishment was necessary to raise children and how they punished their own children. In both countries, participants were also asked whether they observed violence between their parents or in their households during their childhood.
All women, regardless of marital status, were asked questions about intimate partner violence by a current or former intimate partner during the previous year. Indicators of intimate partner violence differed slightly for the two countries. Guatemalan women were asked whether they had been hit, kicked, or slapped; had something thrown at them; had been threatened with being beat up or killed; or had been threatened with a weapon. In El Salvador, women were asked if they had been hit or kicked, threatened with being killed, received an injury or broken bone, or been burned intentionally. For each country, questions on acts of physical violence were combined to calculate the percentage of women who had experienced at least one of these forms of intimate partner physical violence during the past 12 months (23). In both countries, sexual violence was defined as forced sexual intercourse perpetrated by a current or former intimate partner.
For both countries, bivariate and multivariate analyses included independent variables for which there is evidence of an association with physical punishment (1). These independent variables were categorized in the following ways: age group (1519, 2024, 2529, 3034, 3539, and 4049 years for women and men, and 5059 years for men); residence (urban or rural), level of education (Guatemala: none, primary, secondary, postsecondary; El Salvador: none, 13 years, 46 years, 79 years, 10+ years); socioeconomic status (SES)1 (low, medium, high); marital status (single, married or in a consensual union, divorced/widowed/separated), number of live births (none, 12, 34, 5+); and, in Guatemala only, ethnic origin (indigenous, Ladino). Multivariate analyses of the association between intimate partner violence and experience of beating as a child included only women currently married or in union.
For each country, bivariate analyses examined the association between having been beaten as a child and demographic characteristics of participants. For Guatemala, bivariate analyses also examined the association between parental attitudes about the need for physical punishment and the types of punishment used by parents on their own children. Multivariate analyses controlling for key demographic factors were used to examine (1) whether women and men who experienced each type of punishment used this form of punishment on their own children, and (2) whether the experience of beating as a child was associated with intimate partner violence among women who were married or in a consensual union. All analyses were conducted using STATA Version 9.2, applying the sampling weights and controlling for the sampling design. All tables in this analysis show weighted percentages and unweighted numbers of respondents.
Characteristics of participants
Demographic characteristics of the women and men who were asked the questions on violence and thus included in the analyses are shown in Table 1. Because the violence questions were asked only of people who could be interviewed alone (without anyone over age 2 present), those persons who were included in this analysis were slightly less likely to be married and had fewer children than the full sample of women and men interviewed for the survey (not shown).
Punishment experienced as a child
According to the data from Guatemala, 21% of women and 7% of men reported that they were never punished by their parents as children (see Table 2). Among women and men who experienced punishment, the most common type was verbal reprimands, reported significantly more often by men than women. In Guatemala, beating was the second most common form of punishment for both women and men (35% of women, 46% of men, P < 0.01). Smaller proportions of Guatemalan women and men had experienced spanking, with no significant difference by sex. Finally, compared with men in Guatemala more than twice as many women reported having something prohibited (including having something they liked taken away or not being allowing to go out), and the ratio by sex was very similar for having been locked up.
In El Salvador, the rates of any form of punishment as a child were lower, with 44% of women and 24% of men never being punished (P < 0.001). We note, however, that the El Salvadoran sample did not have a precoded option for "spanking" or being "locked out of the house," which may have affected overall reporting. In El Salvador the most common type of punishment reported was beating, a method that men (62%) reported significantly more often than women (42%).
Characteristics of participants who were punished with beating
The characteristics of participants who were beaten as children are shown in Table 3. Findings from both countries suggest a generational effect, with significantly higher percentages of older participants reporting having been punished with beating. The association of beating with age was particularly significant among women in both countries. Difference in experience of beating by classification (urban, rural) of current residence was significant only among Guatemalan men, where urban residents were less likely to have experienced this type of punishment. Higher educational attainment was associated with less likelihood of beating among women from both countries and Guatemalan men. In Guatemala, SES was inversely associated with being beaten as a child for both sexes, but in El Salvador no associations were found with SES. Marital status and number of live births were significantly associated with having been beaten, with greater proportions of ever married or in-union (married, in union, divorced/widowed/separated) women and men from both countries, as well as those with more children, having experienced beating during childhood than never-married women and men and those with fewer children, respectively. In Guatemala, indigenous ethnic origin was significantly associated with having experienced beating in comparison with being a woman of Ladina origin. Ethnicity information was not collected in the El Salvadoran sample.
Association of severe physical punishment and witnessing violence between parents during childhood
The ENSMI 2002 and FESAL 20022003 asked survey participants whether they had witnessed physical violence between their parents during childhood. In Guatemala, 28% of women and 26% of men reported witnessing parental violence; in El Salvador, the comparable proportions were 24% and 20%. The associations between having been beaten as a child and having witnessed physical violence between their parents are shown in Figure 1. Among women in Guatemala who witnessed parental violence, significantly more (49.6%) had been punished with beating as a child than had those who did not witness violence (38.3%). Among Guatemalan men who reported witnessing this type of violence, more than twice as many experienced a beating during childhood as did those who had not. A similar pattern was observed in El Salvador among women (65% and 35%, respectively) and among men (80% and 57%, respectively).
Physical punishment of children by parents interviewed in Guatemala
Among parents of children under age 18 interviewed in Guatemala, 37% of mothers and 32% of fathers reported that they believed physical punishment was necessary to discipline their children (data not shown). In contrast, among parents who reported having used beating to punish their children, 82% of women and 74% of men reported that they believed physical punishment was necessary.
Among Guatemalan survey participants who had children under age 18, 26% of mothers and 23% of fathers reported having never punished their children (thus, 74% and 77% respectively, had used some form of punishment, which includes nonphysical forms) (Table 4). Twenty-six percent of mothers and 20% of fathers reported punishing their children with beating. Table 4 also compares the most commonly reported types of punishment experienced by Guatemalan parents during their own childhood with the types of punishment they reported having used to discipline their own children. The most common form of childhood punishment experienced was verbal, and the proportion of parents reporting use of verbal punishment on their own children was similar. Similar levels of spanking and prohibiting something were also found. In contrast, both mothers and fathers reported less use of beating with their children than they reported having received during their own childhood.
After we controlled for demographic factors determined in our analyses to be associated with use of punishment on children, including age, education, residence, age of children, marital status, and indigenous status, we found that women and men who experienced a specific form of punishment as children were significantly more likely to use that type of punishment on their own children than were women and men who did not experience that punishment as children (Table 5). For example, women who experienced verbal reprimands as a child were 4.3 times (confidence interval [CI], 3.65.1) as likely to have used that form of punishment on their own children as women who did not experience verbal punishment. Spanking had the highest odds ratio, with men and women who were spanked being more than nine times as likely to spank their children as men and women who had not been spanked. Beating also had a high likelihood of being repeated on the child. Mothers and fathers who had not been punished were about four times more likely not to punish their own children than all others.
Experience with punishment during childhood and recent intimate partner violence experienced by women in union
To further explore whether there was a pattern of intergenerational violence, we examined the association between physical punishment during childhood and involvement in a violent relationship with an intimate partner as an adult. Among Guatemalan women who were married or in union at the time of the survey, 11% reported experiencing physical violence or forced sex involving an intimate partner during the last 12 months (the woman's partner threatened to beat her; kill her; use a weapon on her; threw objects at her; hit, kicked, or slapped her; or forced her to have sex). The comparable proportion for El Salvador was 6%. In both Guatemala and El Salvador, a significantly greater percentage of women who were beaten as a child reported recent experience with a violent relationship (Figure 2).
Numerous factors have been found to be associated with an increased risk of intimate partner violence, including current age under 30 years, young age at first marriage, being married for a second time, higher number of live births, lower SES, place of residence, educational status, and history of witnessing familial violence (24). Controlling for these factors using multivariate logistic regression analysis methods, we found that women who were beaten as a child were 1.45 (95% CI, 1.151.84) times more likely in Guatemala and 1.90 (95% CI, 1.342.70) times more likely in El Salvador to currently be in a violent relationship than women who were not beaten as a child (results not shown).
The findings from this study offer insights into the prevalence, types, and intergenerational patterns of child-hood punishment in two Central American countries. In nationally representative samples of women aged 1549 and men aged 1559 in their respective countries, beating was the most common form of punishment experienced during childhood for both men and women in El Salvador, and it was the second most common type in Guatemala. In addition, in both countries, older participants were more likely to have been punished with beating than were younger participants, suggesting a possible decline in beating in recent generations. Whether this is a true decline or it represents under-reporting in recent generations needs to be explored further using similar nationally representative samples from a later period. Further, familial violence was related to childhood beating in both Guatemala and El Salvador, where both men and women who wit-nessed violence between their parents were more likely to have been beaten as children. The experience of physical punishment as a child also had lingering effects in adult relationships, with women who were beaten being more likely to be involved in violent relationships, after controlling for key demographic factors.
These findings are similar to other studies that have examined intergenerational transmission of violence and show that witnessing violence and experiencing abuse in childhood are both associated with a greater likeli-hood of using physical punishment on one's own children in adulthood and a greater likelihood of being in a violent relationship as an adult (3, 16, 20). Notably, most of the previous studies that have demonstrated intergenerational transmission of violence have used small populations, often from a target high-risk setting (e.g., a domestic violence shelter) (20). Our study uses nationally representative data from a large sample of women and men and thus provides stronger evidence of the intergenerational effects of violence in Guatemala and El Salvador.
Our analysis showed that, compared with the adults' generation, fewer offspring are being beaten and a greater percentage of offspring are receiving no punishment in Guatemala, where data on type of punishment used on children were available. This adds further support to the intergenerational decline in beating and punishment overall, as mentioned above. It should be noted, however, that after controlling for associated demographic factors, parents from Guatemala who experienced a particular form of punishment as children were significantly more likely to revisit that severe punishment on their own children than those parents who were not exposed to that form of severe punishment. Similarly, parents who did not experience severe punishment as a child refrained from using severe punishment on their own children, controlling for other factors associated with punishment. Thus, not all parents will pass their childhood punishment experience on to their children (3); however, those parents who beat their children are significantly more likely to have been beaten as children.
This study has several limitations. First, childhood punishment is reported retrospectively, which means that some women and men may not remember the punishment they experienced, leading to underreporting of punishment. Second, the data were collected at the household level, which means there may have been additional reasons for underreporting of punishment experienced (and used) if respondents were worried about breach of confidentiality or were concerned about their and their children's safety if they were forthcoming with information. Third, because punishment was self-reported, some women and men may have chosen not to reveal their experience with beating or another form of punishment; these persons may differ from those persons who choose to reveal their experience. Moreover, in Guatemala, parental reports of punishment used on their own children are perhaps even more susceptible to bias because of a likely unwillingness to report having used harsher forms of punishment on their own children. Similarly, adult women's reports of experience with intimate partner violence are likely under-reported because of concerns about confidentiality or safety. Also, because information on type of punishment used on the respondents' own children was not available in El Salvador, this part of the analysis is limited to the Guatemala samples. Finally, the data are cross-sectional, which makes it difficult to establish the direction of causality between the current demographic factors and punishment experienced in childhood.
Policy and program implications
This analysis reveals that physical punishment of children is a common problem in Guatemala and El Salvador. The farreaching effects of physical punishment warrant the introduction of policy and programmatic interventions to reduce the incidence of childhood beating and other forms of punishment that can lead to long-term negative outcomes. Improving multilevel coordination between many sectors of the communityincluding governmental agencies, courts, nongovernmental organizations, health care providers, parent and youth organizations, schools, and churchescan contribute to the development of a national strategy to protect the youngest members of society from abuse (25).
The 2002 World Report on Violence and Health and other recent studies describe several approaches to reducing the prevalence of harsh physical punishment of children. Educational programs in parenting that provide critical information on children's developmental needs and necessary resources to address those needs have been found to be effective. Counseling programs for parents and caregivers have been found to help parents alleviate stress and provide alternative forms of child discipline other than severe physical punishment (2, 26, 27).
Although most parents in different cultural settings have some understanding of the damaging impact of severe maltreatment of children, acceptable and unacceptable behaviors have to be defined within a cultural context to challenge beliefs that support the use of severe physical punishment. In cultures where the extended family and community have a strong impact, such as in Latin America, it is important to include these social networks in programs that address child punishment (27, 28).
As demonstrated in this study, living in a household where there is physical violence between father and mother is associated with child beating. Moreover, women who were beaten as children may be at greater risk of entering or remaining in a violent relationship than women who were not beaten. Prevention efforts need to address the potential co-occurrence of different types of violence within the family setting. Health professionals need to be trained to identify the warning signs of violence in the home and need to be encouraged to take the time to identify and refer victims of violence to appropriate services (2, 11). With these types of multilevel policy and program changes, there is promise of a reduction in the negative and lingering effects of severe physical punishment for future generations of children in Guatemala and El Salvador.
Acknowledgment. The authors thank Edgar Hidalgo, Jose Mario Caceres, Paul Stupp, Steve McCracken, and Florina Serbanescu for their helpful insights.
1. Orpinas P. Who is violent?: factors associated with aggressive behaviors in Latin America and Spain. Rev Panam Salud Publica. 1999;5(4/5):23244.
2. World Health Organization. World report on violence and health. Geneva: WHO; 2002.
3. Ferrari AM. The impact of culture upon child rearing practices and definitions of maltreatment. Child Abuse Negl. 2001;26(8):796813.
4. Frias-Armenta M. Long-term effects of child punishment on Mexican women: a structural model. Child Abuse Negl. 2002;26(4):37186.
5. Roberts JV. Changing public attitudes towards corporal punishment: the effects of statutory reform in Sweden. Child Abuse Negl. 2000;24(8):102735.
6. American Academy of Pediatrics. Committee on School Health. Corporal punishment in schools. Pediatrics. 2000;106(2 Pt 1):343.
7. International Society for Prevention of Child Abuse and Neglect (ISPCAN). World perspectives on child abuse. 7th ed. Chicago: ISPCAN; 2006.
8. Grogan-Kaylor A. Corporal punishment and the growth trajectory of children's antisocial behavior. Child Maltreat. 2005;10(3):28392.
9. Sadowski LS, Hunter WM, Bangdiwala SI, Munoz SR. The world studies of abuse in the family environment (WorldSAFE): a model of a multinational study of family violence. Inj Control Saf Promot. 2004;11(2):8190.
10. Krugman S, Mata L, Krugman R: Sexual abuse and corporal punishment during child-hood: a pilot retrospective survey of university students in Costa Rica. Pediatrics. 1992;90:15761.
11. Lopez Stewart C, Lara MG, Amighetti LD, Wissow LS, Gutierrez I, Levav I, et al. Parenting and physical punishment: primary care interventions in Latin America. Rev Panam Salud Publica. 2000;8(4):25767.
12. Rodgers CS, Lang AJ, Laffaye C, Satz LE, Dresselhaus TR, Stein MB. The impact of individual forms of childhood maltreatment on health behavior. Child Abuse Negl. 2004;28:57586.
13. Herronkohl RC, Russo MJ. Abusive early child rearing and early childhood aggression. Child Maltreat. 2001;6(1):316.
14. Lee LC, Kotch JB, Cox CE. Child maltreatment in families experiencing domestic violence. Violence Vict. 2004;19(5):57391.
15. Dietz TL. Disciplining children: characteristics associated with the use of corporal punishment. Child Abuse Negl. 2000;24(12):152942.
16. Berger LM. Income, family characteristics, and physical violence toward children. Child Abuse Negl. 2005;29(2):10733.
17. Maker AH, Shah PV, Agha Z. Child physical abuse: prevalence, characteristics, predictors, and beliefs about parent-child violence in South Asian, Middle Eastern, East Asian, and Latina women in the United States. J Interpers Violence. 2005;20(11):140628.
18. Edelson J. The overlap between child maltreatment and woman battering. Violence Against Women. 1999;5(2):13454.
19. Edelson J. Children's witnessing of adult domestic violence. J Interpers Violence. 1999;14(8):83970.
20. Renner LM, Slack KS. Intimate partner violence and child maltreatment: understanding intra- and intergenerational connections. Child Abuse Negl. 2006;30(6):599617.
21. Asociación Demográfica Salvadoreña (ADS) and Centers for Disease Control and Prevention (CDC). Informe final: encuesta nacional de salud familiar FESAL 2002/03. San Salvador: ADS; 2004.
22. Guatemala Encuesta Nacional de Salud Materno Infantil, 2002, Informe Final. Universidad del Valle, CDC Division of Reproductive Health. Vol. 1 Female data; Vol. 2 Male data. Guatemala City: Universidad del Valle de Guatemala; 2003.
23. Speizer I, Goodwin M. Intimate partner violence and unintended pregnancy in El Salvador, Guatemala, and Honduras. CDC research brief. Presented at the Central American Violence Conference, June 2005, Atlanta, GA.
24. Kishor S, Johnson K. Profiling domestic violencea multi-country study. Calverton, MD: ORC Macro; 2004.
25. Garcia-Moreno C, Jansen H, Ellsberg M, Heise L, Watts C. WHO multi-country study on women's health and domestic violence against women. Geneva: WHO; 2005.
26. Smith DE, Mosby G. Jamaican child-rearing practices: the role of corporal punishment. Adolescence. 2003;38(150):36981.
27. Maiter S, Alaggia R, Trocme N. Perceptions of child maltreatment by parents from the Indian subcontinent: challenging myths about culturally based abusive parenting practices. Child Maltreat. 2004;9(3):30924.
28. D'Antonio IJ, Darwish AM, McLean M: Child maltreatment: international perspectives. Matern Child Nurs J. 1993;21(2):3952.
Manuscript received on 16 March 2007.
Revised version accepted for publication 22 January 2008.
* Send correspondence and reprint requests to: Ilene Speizer, University of North Carolina, School of Public Health, Department of Maternal and Child Health, 206 West Franklin Street, Chapel Hill, NC 27516; telephone: 919-966-7411; fax: 919-966-2391; e-mail: Ilene_ email@example.com
1 SES was an unweighted count of the number of household services and durable goods in the woman's home. Included in the SES calculation were the number of rooms in her home, type of cooking fuel used, type of toilet used, whether the home had electricity, and whether the family owned a television.