Multiple victimization of Spanish adolescents: a multilevel analysis.

Lila, Marisol ; Herrero, Juan ; Gracia, Enrique 等

Adolescents are among the most vulnerable to victimization (Hamby & Finkelhor, 2001). Some studies indicate that adolescents are victimized at two or three times the rate of adults, and experience victimization episodes that are equally injurious as those perpetrated against adults (Finkelhor, Ormrod, Turner, & Hamby, 2005a; Wells & Rankin, 1995). Moreover, recent research confirms the pervasive exposure of children and adolescents to violence, crime, maltreatment, and other forms of victimization as a routine part of ordinary childhood. For example, in a study using a United States representative sample of adolescents, Finkelhor et al. (2005a) found that more than half of the participants had experienced a physical assault, more than I to 4 a property victimization, more than 1 in 8 a form of child maltreatment, 1 in 12 a sexual victimization, and more than 1 in 3 had witnessed violence or another form of indirect victimization.

Research on adolescents' victimization has increased, and most of the studies pay attention to prevalence, antecedents and/or consequences of specific forms of victimization (Finkelhor, Ormrod, Turner, & Hamby, 2005a,b; Saunders, 2003). For example, there is a growing number of studies on sexual abuse of adolescents (e.g., Finkelhor & Hashima, 2001; Kendall-Tackett, Williams, & Finkelhor, 1993), bullying (e.g., Duncan, 1999; Nansel, Overpeck, Puilla, Ruan, & Scheidt, 2003), or the witnessing of domestic violence (e.g., Fantuzzo & Mohr, 1999; Kolbo, Blakely, & Engleman, 1996). Likewise, various forms of adolescent victimization have been linked to a multitude of negative outcomes (see Saunders, 2003, for a review). For example, youth victimization has been associated with delinquent behavior (Howing, Wodarski, Lurtz, Gaudin, & Herbst, 1990; Kilpatrick, Saunders, & Smith, 2002; Smith & Thornberry, 1993), problematic substance use (Dembo, Williams, Schmeidler, Berry, Wothke, Getreu et al., 1992; Kilpatrick, Acierno, Saunders, Resnick, Best, & Schnurr, 2000; Kilpatrick, Ruggiero, Acierno, Saunders, Resnick, & Best, 2003), mental health problems (Epstein, Saunders, Kilpatrick, & Resnick, 1998; Kilpatrick, Saunders, & Smith, 2002; Turner, Finkelhor, & Ormrod, 2006), medical and physical problems (Hanson, Davis, Resnick, Saunders, Kilpatrick, & Holmes et al., 2001) suicidality (Bryant & Range, 1995; Saunders, Kilpatrick, Hanson, Resnick, & Walker, 1999), risk of revictimization (Saunders, Kilpatrick, & Resnick, 1998), and other undesirable consequences. As Saunders (2003) pointed out, "It is not uncommon for victims of violence to exhibit certain types of psychiatric disorders or dysfunctional behaviors at rates three, four, and even five times greater than nonvictims" (p. 358).

Additionally, a growing body of research suggests that many of the types of violence are not unique, singular experiences. Rather, it is common for children and adolescents to have been victims of several types of violence on multiple occasions (Green, Goodman, Kupronick, Corcoran, Petty, & Stocton et al., 2000; Kilpatrick & Saunders, 1999; Saunders, 2003). For example, Menard and Huizinga (2001), in their high-risk sample of adolescents, found "chronic multiple victimization" to be the norm. In fact, adolescents who have been exposed to only a single episode of one type of violence are a minority of victimized adolescents (Saunders, 2003). As Finkelhor et al. (2005a) noted, recent research confirms that multiple victimization is common, that different kinds of victimization are interrelated--adolescents who experience one type are also likely to be exposed to other forms of victimization--and that it is more likely that multiple victimized adolescents present distress and other psychopathological symptoms (Finkelhor, Ormrod, & Turner, 2007; Lauritsen & Quinet, 1995; Manion & Wilson, 1995; Outlaw, Ruback, & Britt, 2002). Also, not accounting for multiple victimization forms can make it difficult to identify those adolescents with higher risk for victimization or those victimized chronically (Kochenderfer Ladd & Ladd, 2001). In the study cited above by Finkelhor et al. (2005b), nearly one half of the sample had experienced more than one type of direct or indirect victimization. In these authors' words "this suggests the degree to which studies focusing on a single form of victimization miss a much bigger picture" (p. 18).

Despite this growing body of research on multiple victimization, little attention has been paid to the multiple contexts were multiple victimization may take place, as well as to the different correlates associated with multiple victimization in different contexts. As some research suggests, adolescents and youth may be multiply victimized in more than one context--home, school or the street (see Finkelhor et al., 2007). For example, connections between intrafamily and extra-family victimization have been observed (Baldry, 2003; Perry, Hodges, & Egan, 2001). Nonetheless, with some exceptions (e.g. Appel & Holden, 1998), studies of adolescent victimization have not incorporated these interconnections between contexts. This paper aims to address this gap in the literature by examining correlates of multiple victimization in different contexts (home, school, and the street).

The Present Study

Whereas in other countries the study of adolescent victimization and interest in multiple victimization are increasing, in Spain this kind of research is scarce. Little is known about the incidence of multiple victimization among adolescents in Spain, and almost no research is available on the correlates of multiple victimization in different contexts. The aim of this paper is threefold. First, to explore the incidence of different types of victimization (stealing, hitting, insulting, threatening, blackmailing, and weapon intimidation) in three different contexts (home, school, and street) in a representative sample of Spanish adolescents. Second, to analyze intercorrelations between contexts of multiple victimization. Third, to analyze correlates of multiple victimization in three contexts. Using a multilevel approach, individual and school level correlates of multiple victimization will be analyzed. At the individual (student) level, sociodemographic factors (age, gender) will be explored as they have been found to be related to victimization in adolescents (e.g., Brown & Bzostek, 2003; Finkelhor & Hashima, 2001; Finkelhor et al., 2005a; Herrero, Estevez, & Musitu, 2006; Khoury-Kassabri, Benbenishty, & Astor, 2004; Owens, Daly, & Slee, 2005; Van Dorn, 2004). The quality of family relationships and deviant behavior also will be analyzed, as they have been considered theoretically relevant variables related to child and adolescent victimization (e.g., Gerard & Buehler, 1999; Herrero et al., 2006; Lila, van Asken, Musitu, & Buelga, 2006; Perry, Hodges, & Egan, 2001; Lauritsen, 2003; Lauritsen, Sampson, & Laub, 1991; Shaffer & Ruback, 2002). At the school level, the type of school (public or private) will also be analyzed as it has been associated with the likelihood of being victimized (e.g., Clark & Lab, 2000; Dinkes, Cataldi, Kena, & Baum, 2006; Lab & Clark, 1997; U.S. Department of Education, 1993).

METHOD

Participants

We used data from 1,908 adolescents 13-18 years old of the Valencian Community (Spain). Multi-stage sampling procedures were followed to obtain a representative sample of adolescents studying at the time of the survey (year 2002). In the first stage, a random selection of public and private schools in the Valencian Community was carried out. At this stage, 39 schools were selected (20 public and 19 private). Principals of these schools were invited to participate in the study. Although the level of commitment to the study was high among principals, some of them conditioned their participation to examination of the questionnaire prior to its administration. Others directly asked for the results of the study once they were available. In all cases, researchers complied with these requirements.

In the second stage, a random selection of adolescents proportional to students in each school was performed. Letters with a brief description of the program and explaining the need to collect data were sent to families. This letter also included a no-consent form if parents did not wish their children to take part in the program. No students returned the no-consent form. Finally, trained personnel administered the questionnaires in each classroom according to the schedule agreed upon by principals and researchers. Students completed questionnaires in a regular class period (1 hour) with only the trained personnel present.

The final sample consisted of 1,908 adolescents 13-18 years old (M = 14.51, SD = 1.20) of whom 50.52% were boys. Due to the larger size of public schools, 54.6% of the sample were from this type of school.

Outcome variables

Multiple victimization. Self-reported multiple victimization was measured by asking students whether they had experienced any of six situations in the street, at school, and at home (see Table 1). Three measures of multiple victimization (range 0-6) were obtained summing up the number of situations experienced in each context: multiple victimization in the street, multiple victimization at school, and multiple victimization at home. Means and standard deviations for these measures of multiple victimization are presented in Table 1.

Insults (46.3%) and thefts (32.1%) were the most frequent victimizations experienced in the street. A majority of students indicated they had been insulted at school (63.4%)--clearly the most common type of victimization in this context. Regarding the home context, the percentage of situations experienced sharply decreased as compared to the other two contexts.

Overall, adolescents experienced more types of victimization in the street than at school and, at home. Victimization in the street (1.39) and at school (1.22) double the level of victimization at home (0.53).

Only 10.8% (N = 207) of the sample noted no victimization at all in all of the contexts analyzed; 22.7% (N = 434) experienced some victimization in the three contexts; 35.6% (N = 680) experienced victimization simultaneously in two contexts, and 30.8% (N = 587) in only one context.

Student-level Variables

Family relationships. Participants were asked to estimate the quality of relationships with family members: In general, how is your relationship with persons in the household? Possible responses were: 1 = very bad, 2 = bad, 3 = neither bad nor good, 4--good, and 5 = very good. (M = 4.21, SD = 0.75).

Deviant behavior. This is a global concept that reflects behavior that differs from accepted standards and bring disapproval; it comprises both antisocial behavior (theft, vandalism, and damage to property), aggression (verbal and physical), and substance use (see Herrero, Estevez, & Musitu, 2006). A 33-item checklist was used to measure deviant behavior. Items in this list were adapted from Emler and Reicher's (1995) self-reported delinquencies checklist, that measures deviant behavior such as theft (e.g., stole from a large store while it was open), vandalism (e.g., broke windows of empty houses), aggression (attacked an enemy in a public space), or drugs (e.g., took illegal drugs), (M = 7.56, SD = 6.17).

Sociodemographic variables. Age (in years) and gender (1 = boy, 2 = girl) were used as the variables.

Type of school was coded 1 for public school, and 2 for private school.

ANALYTIC STRATEGY

Date present a hierarchical structure with students (level 1) nested within schools (level 2). Multilevel modeling was used to allow for inclusion of additional error terms that reflect the complex pattern of variation introduced by the hierarchical structure of the data (random effects) (Raudenbusch & Bryk, 2002). Since our interest was in analyzing the associations of the covariates of the study with the three outcome variables (three contexts of multiple victimization) we estimated multivariate multilevel models using the HMLM2 module of the statistical package HLM 6.02. Multivariate multilevel modeling allows for estimation of the association of each covariate with a set of intercorrelated outcome variables, while accounting for the hierarchical structure of the data. This procedure has several advantages as compared to analysis of three separate regression equations for each outcome variable. Among them, it allows for statistical tests of the difference in magnitude for each of the regression coefficients across outcome variables. For instance, it allows us to answer such questions as: Is the association of age and multiple victimization the same for all contexts of victimization (street, school, and home)?

We checked for multicollinearity problems among predictors examining the Variance Inflation Factor (VIF), all off-diagonal elements in the variance-covariance (Tau) matrix for correlations close to 1 or -1, and the diagonal elements for any elements close to zero, with no indication of multicollinearity.

The multilevel analysis was performed in two steps. The starting point was an empty model without explanatory variables in which the total variance of victimization was partitioned into a component at each level. This model (empty model) was used to test for random variation of the outcome variables at different levels and, consequently, if a multilevel approach was reasonable. In the second step we explored fixed effects of variables at the student and school level and tested for significant differences of coefficients across the three outcome variables.

RESULTS

Table 2 presents zero-order correlations among the outcomes variables and the zero-order correlations with the student and school level covariates of the study.

As for correlations among the outcome variables and both student and school level covariates, Table 2 shows a different pattern for each context of victimization. Thus, while some covariates are highly correlated with all contexts of victimization (family relationships), others are correlated only with some contexts of victimization (gender, age, deviant behavior, and type of school). Results noted in Table 2 suggest that those adolescents who experience victimization in one context tend also to experience victimization in other contexts. These results, however, do not take into account the hierarchical structure of the data, and they also do not provide statistical significance of the strength of association of each covariate with each context of victimization relative to other contexts.

Results shown in Table 3 take into account these circumstances and allow us to compare the relative strength of the association of each covariate with the three contexts of victimization while also taking into account the hierarchical structure of the data. At the bottom of Table 3 random variation is shown of the outcome variables at the student and school level for both the empty and final model. In first examining the results for the empty model (with no predictors), we find random variation of victimization both between students and schools. This suggests that we need to control for the hierarchical structure of the data (multilevel approach). Adding all covariates reduces random variation both at the student and school level. For the student level, the reduction is 42% (0.29/0.40 + 0.29 = 0.42). For the school level it is 33%

In examining the statistical significance of random effects for the model presented in Table 3, we can conclude that although the covariates substantially reduced random variation at the student level, there is still some variation that the model cannot account for, and that the inclusion of other relevant student-level covariates would be appropriate. For the school level, we note that no random variation is left by the model, suggesting that both student-level and school-level covariates account for all the between-schools variations. In other words, once we controlled for characteristics of students within school and type of school, all of the sampled schools showed similar levels of victimization.

Since parameters in Table 3 are centered around the grand mean, the intercept may be interpreted as the level of victimization of an average student of the sample. Adding (or subtracting) statistically significant parameters to the intercept give us the estimated level of victimization for specific sociodemographic groups and/or specific levels of other covariates. For instance, the average student in this sample experienced 1.38 situations of victimization in the street. An 18-year-old boy with poor family relationships, who is highly engaged in delinquent behaviors will experience an estimated 4.12 situations in the street. The maximum level of victimization at school would be for a boy with the poorest family relationships and the highest level of deviant behavior (estimated at 2.06). Likewise, adolescents with the poorest family relationships and higher levels of delinquency, mainly students attending private schools, would experience an estimated 1.54 situations of victimization at home.

An important feature of multivariate multilevel models is the possibility of comparing the strength of association of the covariates with each of the outcome variables, once we take into account that these outcome variables are intercorrelated (see note in Table 3). Table 3 also shows that some covariates are statistically associated with all different outcomes while others are associated only with some of the outcome variables. This had been forecasted in Table 2, but the multilevel results of Table 3 allow us to formally test the multivariate associations among covariates and outcomes.

When we take into account the adolescent scores in all the covariates of the study, we see that the association of age with multiple victimization at home is no longer significant. Family relationships are a significant covariate of all contexts of multiple victimization, suggesting that victimized adolescents have poorer quality family relationships. When testing for the statistical significance of these associations across different contexts of multiple victimization, we can conclude that those victimized at home have the strongest association, while those victimized at school or in the street have similar associations. Deviant behavior is also a significant correlate of all contexts of multiple victimization. Adolescents who engage in more deviant behaviors are also more victimized in the three contexts analyzed in this study (street, school, and home). This positive association is greater for victimization in the street.

As for the covariates that are related only to some contexts of victimization, gender is associated with victimization in the street and at school. Girls have lower levels of victimization in these contexts than do boys, although both genders present similar levels of home victimization. Also, the gender-victimization relationship is greater for multiple victimization in the street. The average boy will experience 1.38 experiences of victimization in the street whereas an average girl will experience 0.86 (1.38-0.52 = 0.86). This means that average girls are 48% less exposed to victimization experiences in the street than are boys. For victimization at school, this percentage decreases sharply to 15%.

As to age, there is a positive association with victimization in the street but not with the remaining two contexts of victimization (school and home). The older participants of the study experience more victimization in the street but their level of victimization at school and at home remained similar to that of, the younger participants. We also found that students from private schools experienced more victimization at home than those attending public schools (20% more), but their levels in other contexts of victimization equaled those of adolescents attending public schools.

DISCUSSION

The present study aimed to explore multiple victimization in a representative sample of 1,908 students 13-18 years old living in the Valencian Community, Spain. First we analyzed the incidence of different types of victimization (stealing, hitting, insulting, threatening, blackmailing, and weapon intimidation) in three different contexts (home, school, and street). We found that multiple victimization in the street was the most frequent, followed by school, and finally, with the lowest levels in the sample, at home. Among the students analyzed, the number of victimization experiences in the street and at school was twice the number of victimization experienced at home. We also found that only a minority (10.8%) of adolescents were not victimized in any context, and that most respondents experienced victimization at least in two (36.6%) or three (22.7%) of the contexts. These results support other research conducted in different countries indicating that multiple victimization in adolescence appears to be the rule rather than the exception (Green, Goodman, Kupronick, Corcoran, Petty, & Stocton et al., 2000; Kilpatrick & Saunders, 1999; Menard & Huizinga, 2001; Saunders, 2003). As several scholars have pointed out (e.g., Green et al., 2000; Kilkpatrick & Saunders, 1999; Saunders, 2003), multiple victimization in adolescence is not uncommon and adolescents who experience one type of victimization are likely to be exposed to other types. For example, Finkelhor et al. (2007) observed that previously victimized children, particularly those experiencing child maltreatment or family violence, also appear to be at greater risk of subsequent victimizations (Duncan, 1999a,b). Moreover, when observed over time in schools, some children appear to be chronically targeted year aider year (Perry et al., 2001). In this respect, we also found that the three contexts of multiple victimization were intercorrelated, suggesting that those adolescents who experience multiple victimization in one context tend also to experience multiple victimization in other contexts (Appel & Holden, 1998; Finkelhor et al., 2007). To account for the fact that some adolescents tend to be victimized in different contexts, Finkelhor et al. (2007) suggest that the clustering of victimization may be explained by a number of factors that increase the risk; e.g., families and neighborhoods (Cicchetti & Lynch, 1993), characteristics of children, such as pre-existing psychological problems or poor social interactional skills (e.g., Finkelhor, Ormrod, & Turner, in press; Perry et al., 2001; Tseloni & Pease, 2003), poor supervision, or social isolation (e.g., Korbin, 2003; Gracia & Musitu, 2003).

In this study we also analyzed correlates of multiple victimization using a multivariate multilevel regression approach. This allowed us to control both for the intercorrelation of the outcome variables (multiple victimization in the street, at school, and at home) and the hierarchical structure of the data (students nested within schools). This also allowed testing for the significantly statistical differences of the strength of each covariate with each context of victimization. Results indicated that some covariates were significantly associated with victimization in all contexts (family relationships and deviant behavior) while other covariates were statistically related to only some contexts of victimization (gender, age, and type of school).

In relation to gender, we found that this variable was related to victimization in the street and at school. Boys were more victimized than girls in this context. These results are consistent with previous research showing a greater victimization risk for boys (e.g., Brown & Bzostek, 2003; Finkelhor & Hashima, 2001; Finkelhor et al., 2005b; Herrero et al., 2006). We also found age-related differences in victimization, but only for multiple victimization in the streets. Our findings show that multiple victimization in the streets is more common among older adolescents. A possible explanation is that as children get older they acquire greater autonomy and use the streets without parental supervision, which in turn makes them more vulnerable to victimization (Bilchik, 1999). In this respect, we agree with Finkelhor et al. (2007) in that future research should further examine whether multiple victimization is a condition that is more difficult to escape at older ages. Adolescents reporting poorer quality of family relationships also reported higher levels of victimization in each context. These results are consistent with previous studies showing a relationship between a negative family environment and a greater risk of victimization (Gerard & Buchler, 1999; Smith, Bowers, Binney, & Cowie, 1993). According to Perry, Hodges and Egan (2001) a possible explanation of this relationship is that negative family relationships set up internalized cognitive "victim schemas" in some children, that may lead adolescents to behave in a manner that make them more vulnerable to victimization by peers. Deviant behavior was also positively associated with all types of victimization, although this association was greater in the street. A substantial body of research reports a positive association between deviant behavior and victimization (see Herrero et al., 2006, for a review).

Finally, we found a relationship only between type of school and multiple victimization at home (not in the schools or the streets), with multiple victimization at home being greater for students of private schools. This finding is somewhat surprising as results of previous studies found that adolescents who attended public schools had more knowledge of and experience with crime and threats than did those in private schools (e.g., De Voe, Peter, Noonan, Snyder, & Baum, 2005). This is an issue that deserves further study.

Research on multiple victimization in different contexts also has implications for future research on basic issues such as its impact on the psychological and social adjustment of adolescents. As Finkelhor et al. (2005a) suggested, the total number of different victimizations is a more important predictor of negative outcomes than the presence of any particular type of victimization. However, most studies do not control for the effects of other types of victimization when evaluating the potential impact of a particular type of violence (Rossman & Rosenberg, 1998; Saunders, 2003). Not taking into account multiple victimization leads to problems such as the underestimation of the number and diversity of adolescents' victimization or to difficulty in understanding the relationship among different victimization forms (Finkelhor et al., 2005b). Outcomes apparently associated with one type of violence might well be the result of another type of violence, the cumulative result of exposure to multiple types of violence, and/or a complex interaction of violence types (Saunders, 2003). Furthermore, adolescents victimized in different ways and in different contexts might be more affected than children repeatedly victimized by just one person or in just one context (Cohen, Perel, DeBellis, Friedman, & Putnam, 2002; Finkelhor, Ormrod, & Turner, 2007). In this regard Finkelhor et al. (2007) suggest that it is important for practitioners to identify children who have experienced multiple victimizations: "Because of their higher levels of traumatic symptomatology, poly-victims may merit priority attention" (p. 20).

Overall, our findings suggest the need for a more multidimensional approach to adolescent victimization than the one that has been characteristic of the field to date. Researchers and policy makers concerned about victimization in adolescence need to be more aware of the multiple contexts of victimization, its mutual connections, and its correlates. This multidimensional approach would more importantly yield benefits for victimized adolescents by suggesting intervention and prevention strategies.

REFERENCES

Appel, A., & Holden, G. W. (1998). The co-occurrence of spouse and physical child abuse: A review and appraisal. Journal of Family Psychology, 12(4), 578-599.

Baldry, A. C. (2003). Bullying in schools and exposure to domestic violence. Child Abuse & Neglect, 27, 713-732.

Bilchik, S. (1999). Violence after school. Juvenile Justice Bulletin, November (NJC 178992).

Brown, B. V., & Bzostek, S. (2003). Violence in the lives of children. Child Trends DataBank: Cross Currents, 1, 1-13.

Bryant, S., & Range, L. M. (1995). Suicidality in college women who were sexually and physically punished by parents. Violence and Victims, 10, 195-201.

Cicchetti, D., & Lynch, M. (1993). Toward an ecological/transactional model of community violence and child maltreatment: Consequences for children's development. Psychiatry, 56, 96-118.

Clark, R. D., & Lab, S. P. (2000). Community characteristics and in-school criminal victimization. Journal of Criminal Justice, 28(1), 33-42.

Cohen, J. A., Perel, J. M., DeBellis, M. D., Friedman, M. J., & Putnam, F. W. (2002). Treating traumatized children: Clinical implications of the psychobiology of posttraumatic stress disorder. Trauma Violence & Abuse, 3(2), 91-108.

Dembo, R., Williams, L., Schmeidler, J., Berry, E., Wothke, W., & Getreu, A. et al. (1992). A structural model examining the relationship between physical child abuse, sexual victimization and marijuana/hashish use in delinquent youth: A longitudinal study. Violence and Victims, 7, 41-62.

DeVoe, J. F., Peter, K., Noonan, M., Snyder, T. D., & Baum, K. (2005). Indicators of School Crime and Safety: 2005 (NCES 2006-001/NCJ210697). U.S. Departments of Education and Justice. Washington, DC: National Center for Education Statistics.

Dinkes, R., Cataldi, E. F., Kena, G., Baum, K., & Zinder, T. D. (2006). Indicators of School Crime and Safety, 2006 (NCJ 214262). Justice Department's Bureau of Justice Statistics (BJS) and the Department of Education's National Center for Education Statistics.

Duncan, R. D. (1999). Peer and sibling aggression: An investigation of intra- and extra-familial bullying. Journal of Interpersonal Violence, 14, 871-886.

Emler, N., & Reicher, S. (1995). Adolescence and delinquency: The collective management of reputation. Oxford: Blackwell.

Epstein, J. N., Saunders, B. E., Kilpatrick, D. G., & Resnick, H. S. (1998). PTSD as a mediator between childhood rape and alcohol use among women. Child Abuse & Neglect, 22, 223-234.

Fantuzzo, J., & Mohr, W. (1999). Prevalence and effects of child exposure to domestic violence. The Future of Children, 9(3), 21-32.

Finkelhor, D., & Hashima, P. (2001). The victimization of children and youth: A comprehensive overview. In S. O. White (Ed.), Handbook of youth and justice (pp. 49-78). New York: Kluwer Academic/Plenum.

Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2007). Poly-victimization: A neglected component in child victimization trauma. Child Abuse & Neglect, 31, 7-26.

Finkelhor, D., Ormrod, R. K., & Turner, H. A. (in press). Revictimization patterns in a national logitudinal sample of children and youth. Child Abuse & Neglect.

Finkelhor, D., Ormrod, R. K., Turner, H. A., & Hamby, S. L. (2005a). The victimization of children and youth: A comprehensive, national survey. Child Maltreatment, 10, 5-25.

Finkelhor, D., Ormrod, R. K., Turner, H. A., & Hamby, S. L. (2005b). Measuring poly-victimization using the Juvenile Victimization Questionnaire. Child Abuse & Neglect, 29, 1297-1312.

Gerard, L. M., & Buehler, C. (1999). Multiple risk factors in the family environment and youth problem behaviors. Journal of Marriage and the Family, 61, 343-361.

Gracia, E., & Musitu, G. (2003). Social isolation from communities and child maltreatment: A cross-cultural comparison. Child Abuse & Neglect, 27, 153-168.

Greeen, B. L., Goodman, L. A., Kupronick, J. L., Corcoran, C. B., Petty, R. M., & Stocton, P. et al. (2000). Outcomes of single versus multiple trauma exposure in a screening sample. Journal of Traumatic Stress, 13, 271-286.

Hamby, S. L., & Finkelhor, D. (2001). Choosing and using child victimization questionnaires. Juvenile Justice Bulletin, March.

Hanson, R. F., Davis, J. L., Resnick, H. S., Saunders, B. E., Kilpatrick, D. G., & Holmes, M. et al. (2001). Predictors of medical examinations following child and adolescent rapes in a national sample of women. Child Maltreatment, 6, 250-259.

Herrero, J., Estevez, E., & Musitu, G. (2006). The relationships of adolescent school-related deviant behavior and victimization with psychological distress: Testing a general model of the mediational role of parents and teachers across groups of gender and age. Journal of Adolescence, 29, 671-690.

Howing, P. T., Wodarski, J. S., Lurtz, P. D., Gaudin, J. M., & Herbst, E. N. (1990). Child abuse and delinquency: The empirial and theoretical links. Social Work, 26, 244-249.

Kendall-Tackett, K. A., Williams, L. M., & Finkelhor, D. (1993). Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin, 113, 164-180.

Khoury-Kassabri, M., Benbenishty, R., & Aster, R. (2004). The contributions of community, family, and school variables to student victimization. American Journal of Community Psychology, 34, 187-204.

Kilpatrick, D. G., & Saunders, B. E. (1999). Prevalence and consequences of child victimization: Results from the National Survey of Adolescents: Final report. Charleston, SC: Author.

Kilpatrick, D. G., Acierno, R., Saunders, B. E., Resnick, H. S., Best, C. L., & Schnurr, P. P. (2000). Risk factors for adolescent substance abuse and dependence: Data from a national sample. Journal of Consulting and Clinical Psychology, 68, 19-30.

Kilpatrick, D. G., Ruggiero, K. J., Acierno, R., Saunders, B. E., Resnick, H. S., & Best, C. L. (2003). Violence and risk of PTSD, major depression, substance abuse/dependence, and comorbidity: Results from the national survey of adolescents. Journal of Consulting and Clinical Psychology, 71, 692-700.

Kilpatrick, D. G., Saunders, B. E., & Smith, D. W. (2002). Research in brief: Youth victimization: Prevalence and implications (NCJ 194972). Washington, DC: U.S. Department of Justice, National Institute of Justice.

Kochenderfer Ladd, B., & Ladd, G. W. (2001). Variations in peer victimization: Relations to children's maladjustment. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 25-48). New York: Guilford.

Kolbo, J. R., Blakely, E. H., & Engleman, D. (1996). Children who witness domestic violence: A review of empirical literature. Journal of Interpersonal Violence, 11(2), 281-293.

Korbin, J. E. (2003). Neighborhood and community connectedness in child maltreatment research. Child Abuse & Neglect, 27, 137-140.

Lab, S. P., & Clark, R. D. (1997). Discipline, Control and School Crime: Identifying Effective Intervention Strategies. Washington, DC: National Institute of Justice.

Lauritsen, J. L. (2003). How families and communities influence youth victimization (Juvenile Justice Bulletin). Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

Lauritsen, J. L., & Quinet, K. F. D. (1995). Repeat victimization among adolescents and young adults. Journal of Quantitative Criminology, 11(2), 143-166.

Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). The link between offending and victimization among adolescents. Criminology, 29(2), 265-292.

Lila, M. S., van Aken, M., Musitu, G., & Buelga, S. (2006). Family and adolescence. In S. Jackson & L. Goosens (Eds.), Handbook of adolescent development (pp. 154-174). Hove: Psychology Press.

Manion, I. G., & Wilson, S. K. (1995). An examination of the association between histories of maltreatment and adolescent risk behaviors. Ontario:

Minister of National Health and Welfare.

Menard, S., & Huizinga, D. (2001). Repeat victimization in a high-risk neighborhood sample of adolescents. Youth & Society, 32, 447-472.

Nansel, T. R., Overpeck, M. D., Pilla, R. S., Ruan, W. J., & Scheidt, P. C. (2003). Relationships between bullying and violence among US youth. Archives of Pediatrics and Adolescent Medicine, 157(4), 348-353.

Outlaw, M., Ruback, B., & Britt, C. (2002). Repeat and multiple victimizations: The role of individual and contextual factors. Violence and Victims, 17(2), 187-204.

Owens, L., Daly, A., & Slee, P. (2005). Sex and age differences in victimization and conflict resolution among adolescents in a South Australia School. Aggressive Behavior, 31, 1-12.

Perry, D. G., Hodges, E. V. E., & Egan, S. K. (2001). Determinants of chronic victimization by peers: A review and new model of family influence. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 73-104). New York: Guilford Press.

Raudenbusch, S., & Bryk, A. (2002). Hierarchical linear and nonlinear models: Applications and data analysis methods. London: Sage.

Rossman, B. B. R., & Rosenberg, M. S. (1998). Maltreated adolescents: Victims caught between childhood and adulthood. Journal of Aggression, Maltreatment & Trauma, 2(1), 107-129.

Saunders, B. E. (2003). Understanding children exposed to violence: Toward an integration of overlapping fields. Journal of Interpersonal Violence, 18(4), 356-376.

Saunders, B. E., Kilpatrick, D. G., & Resnick, H. S. (1998). Victim, incident, and offender characteristics of child rape as risk factors for adult rape among women: Results from the National Women's Study. Paper presented at the annual meeting of the International Society for Traumatic Stress Studies, Washington, DC.

Saunders, B. E., Kilpatrick, D. G., Hanson, R. F., Resnick, H. S., & Walker, M. E. (1999). Prevalence, case characteristics, and long-term psychological correlates of child rape among women: A national survey. Child Maltreatment, 4, 187-200.

Shaffer, J. N., & Ruback, R. B. (2002). Violent victimization as a risk factor for violent offending among juveniles. Juvenile Justice Bulletin, December (NCJ 195737).

Smith, P. K., Bowers, L., Binney, V., & Cowie, H. (1993). Relationships of children involved in bully/victim problems at school. En S. Duck (Ed.), Learning about relationships, understanding relationship processes (pp. 184-212). Londres: Sage Publications.

Smith, C. A., & Thornberry, T. P. (1993). The relationship between childhood maltreatment and adolescent involvement in delinquency. Paper presented at the annual conference of Society for Research in Child Development. New Orleans, LA.

Tseloni, A., & Pease, K. (2003). Repeat personal victimization: "Boosts" or "flags"? British Journal of Criminology, 43, 196-212.

Turner, H. A., Finkelhor, D., & Ormrod, R. (2006). The effect of lifetime victimization on the mental health of children and adolescents. Social Science & Medicine, 62, 13-27.

U.S. Department of Education, National Center for Education Statistics (1993). National Household Education Survey.

Van Dorn, R. (2004). Correlates of violent and non-violent victimization in a sample of public high school students. Violence & Victims, 19, 303-320.

Wells, L. E., & Rankin, J. H. (1995). Juvenile victimization: Convergent validation of alternative measurements. Journal of Research in Crime and Delinquency, 32(3), 287-307.

Juan Herrero. Department of Psychology, University of Oviedo, Oviedo, Spain.

Enrique Gracia. Department of Social Psychology, University of Valencia, Valencia, Spain.

Requests for reprints should be sent to Marisol Lila, Department de Psicologia Social, Facultad de Psicologia, Avd. Blasco Ibanez, 21, 46010-Valencia, Spain. E-mail: [email protected]

Table 1. Percentage of students by type and context of multiple
victimization. Means and standard deviations for victimization in
the street, at school and at home (n = 1908).

 Contexts

 Type of Street School
 victimization

Thefts 32.1 12.0
Hits 16.7 18.4
Insults 46.3 63.6
Threats 22.8 15.1
Blackmails 8.2 10.5
Intimidation with weapon 11.2 1.0
Multiple Victimization (1) 1.39 (1.41) (a) 1.22 (1.16) (b)
 [Mean (S.D.)]

 Contexts

 Type of Home
 victimization

Thefts 6.2
Hits 10.0
Insults 18.8
Threats 4.0
Blackmails 11.8
Intimidation with weapon 0.5
Multiple Victimization (1) 0.53 (0.92)
 [Mean (S.D.)]

(1) Number of types of victimization experienced by individuals
(a) > (b) > (c), p <.001

Table 2. Correlations of student and school-level covariates of the
study with multiple victimization (street, school, and home) and
correlations among outcome variables (n = 1908).

 Contexts

 Street School Home

Girls -0.25 *** -0.11 *** -0.02
Age 0.18 *** -0.02 0.08 ***
Quality of family relationships -0.15 *** -0.10 *** -0.28 ***
Deviant behavior 0.09 *** 0.03 0.06 *
Private Schools 0.01 0.02 0.08 **
Correlations among
 outcome variables
School 0.31 *** --
Home 0.28 *** 0.34 *** --

a > b > c, p <.001

* p < .05; ** p < .01; *** p < .001

Table 3. Results from multivariate multilevel regression analysis
of correlates of different contexts of adolescent multiple
victimization (n = 1908). Parameters are unstandardized. All
covariates are centered around the grand mean. (1)

 Contexts

 Street School

Intercept 1.38 (a) 1.22 (b)
Student-level
 Girls -0.52 *** (a) -0.18 *** (b)
 Age 0.12 *** (a) -0.03 (b)
 Family relationships -0.17 *** (b) -0.12 *** (b)
 Deviant behavior 0.07 *** (a) 0.04 *** (b)
School-level
 Private Schools -0.06 (b) -0.01 (b)
Random effects
 Student-level 0.29(0.02) ***
 School-level 0.01
(0.01)*
Model deviance 16.400

 Contexts

 Home

Intercept 0.54 (c) a > b > c, p < .001
Student-level
 Girls -0.06 a > b, p < .001
 Age -0.01 (b) a > b, p < .001
 Family relationships -0.27 *** (a) a > b, p < .001
 Deviant behavior 0.03 *** (b) a > b, p < .001
School-level
 Private Schools 0.11 * (a) a < b, p < .001
Random effects
 Student-level empty model = 0.40 (0.02) ***
 School-level empty model = 0.02 (0.01) *
Model deviance empty model = 17.007

(1) Correlations among outcome variables are: street-school = 0.25,
street-home =0.20, school-home = 0.29, all significant at p < .001
* p < .05; ** p < .01; *** p < .001