For decades, research on intimate partner violence has focused mainly on relationships between adult couples or within the marital domain. However, in recent years, more attention is being paid to the study of the romantic relationships between young couples (Fernández-Fuertes, Orgaz, & Fuertes, Reference Fernández-Fuertes, Orgaz and Fuertes2011; O’Leary, Slep, Avery-Leaf, & Cascardi, Reference O’Leary, Slep, Avery-Leaf and Cascardi2008; Rey-Anacona, Mateus-Cubides, & Bayona-Arévalo, Reference Rey-Anacona, Mateus-Cubides and Bayona-Arévalo2010).
Dating violence (DV) is considered as any type of intentional aggression (physical, mental or sexual) of one member of the couple against the other during dating (Public Health Agency of Canada, 2006). This definition does not determine the age of the people involved, considering it as a term that refers to a specific type of violence called adolescent intimate partner violence, also known as dating violence, violence among young people, etc. (Pazos, Oliva, & Hernando, Reference Pazos, Oliva and Hernando2014). Other authors have defined this type of perpetration of violence as an act where one person hurts another person in the context of a relationship in which there is attraction (Close, Reference Close2005). Coexistence (living together), children and economic independence constitute some of the differences between young couples and adult couples (Viejo, Reference Viejo2014).
In addition, DV can be indirect (threat and emotional verbal aggression) and direct (physical aggression). According to Wolfe, Scott, Wekerle, and Pittman (Reference Wolfe, Scott, Wekerle and Pittman2001), adolescence is the stage in which indirect attacks are more frequent. Other authors have considered indirect violence as relationship aggression, where social isolation of the victim is generated with respect to their peers (van Geel, Vedder, & Tanilon, Reference van Geel, Vedder and Tanilon2014).
When analyzing the prevalence of DV, it is essential to take into account the sample studied, the instruments and methodology used, and the type of violence explored. The present study focuses on adolescents under residential care, considering that they deserve special attention as they are minors with a special casuistry: Without hope of return to the family home, with mental health problems, violent behaviors, offenders with protective measures, unaccompanied foreign minors (UFMs) or accompanied foreign minors under inadequate educational models who often suffer serious emotional or behavioral problems (Bravo & Del Valle, Reference Bravo and Del Valle2009). Only two studies have been found that have explored DV in adolescents under residential care. The first known study was that of Jonson-Reid and Bivens (Reference Jonson-Reid and Bivens1999), resulting from a survey (adaptation of the Bergman questionnaire, 1992) carried out with adolescents under residential care who were taking part in a DV information day. The total number of adolescents who completed the survey was 85, of which 48% reported being a perpetrator and/or a DV victim. Later, Jonson-Reid, Scott, McMillen, and Edmond (2007), using the same survey, found that 8% of their sample (sample composed of 339 17-year-old American adolescents under residential care in the process of emancipation) recognized using violence (physical or psychological) in their dating relationships and 17% having been a victim. The authors explain the discrepancy in the results of these two studies as Jonson-Reid and Bivens’ (Reference Jonson-Reid and Bivens1999) study was performed after an informative talk on DV, which could have conditioned the responses of the participants.
The rest of the studies found focused on samples of adolescents from schools. Haynie et al. (Reference Haynie, Farhat, Brooks-Russell, Wang, Barbieri and Iannotti2013) analyzed both the victimization and perpetration of DV in a representative sample of US adolescents (2,203 students from 80 schools), through the Revised Conflict Tactics Scale (Straus, Hamby, Boney-McCoy, & Sugarman, Reference Straus, Hamby, Boney-McCoy and Sugarman1996). They found a 24% of verbal victimization, a 11% of physical victimization, and similar figures for violence perpetration: a 21% of perpetration of verbal violence and 9% of perpetration of physical violence. According to the results of the 2013 National Survey on risk behavior of young Americans, 20.9% of adolescent girls and 10.4% of adolescent boys had suffered some type of DV in the past 12 months (Vagi, Olsen, Basile, & Vivolo- Kantor, Reference Vagi, Olsen, Basile and Vivolo-Kantor2015). Specifically, among girls, the prevalence of physical victimization was 6.6%. In the case of boys, the prevalence for physical victimization was 4.1%, 2.9% for sexual and 3.3% for both types of victimization.
Studies such as that of Holt and Espelage (Reference Holt and Espelage2005), also with a sample of American adolescent students and using the Victimization in Dating Relationships Scale (Foshee et al., Reference Foshee, Linder, Bauman, Langwick, Arriaga, Heath and Bangdiwala1996) and the Abusive Behavior Inventory (ABI, Shepard & Campbell, Reference Shepard and Campbell1992), found significantly higher victimization figures: A 37% reported having suffered physical violence and 62% reported psychological violence during the past year. Psychological violence perpetration figures are even higher in other studies, such as that of Jackson, Cram, and Seymour (Reference Jackson, Cram and Seymour2000), in which through an ad hoc survey with New Zealand adolescents, found that 81.5% of girls and 76.3% of boys reported having suffered psychological violence (monopolization, degradation or isolation) in their dating relationships. Foshee and Matthew (Reference Foshee, Matthew, Flannery, Vazonsyi and Waldman2007), on the other hand, conducted a review of studies that analyzed perpetration of DV and found prevalence figures of between 14% and 81% for the perpetration of psychological violence, and between 11% and 41% for the perpetration of physical violence.
Although many studies focus only victimization, and not the perpetration of violence (Foshee & Matthew, Reference Foshee, Matthew, Flannery, Vazonsyi and Waldman2007), the research carried out seems to indicate that the figures of victimization and perpetration of violence are similar, and that the prevalence is greater in the case of psychological violence (perpetration and victimization) rather than physical violence, as would be expected (Carrascosa, Cava, & Buelga, Reference Carrascosa, Cava and Buelga2015; Haynie et al., Reference Haynie, Farhat, Brooks-Russell, Wang, Barbieri and Iannotti2013).
Regarding studies conducted with Spanish samples, Muñoz-Rivas, Graña, O’Leary, and González (Reference Muñoz-Rivas, Graña, O’Leary and González2007), using the Modified Conflict Tactics Scale (MCTS) (Neidig, Reference Neidig1986), found that more than 90% of adolescent students in their sample admitted having resorted to the perpetration of verbal violence within their dating relationships, a percentage that drops to 18% for the perpetration of physical violence. These prevalence figures are similar to those found by Fernández-Fuertes and Fuertes (Reference Fernández-Fuertes and Fuertes2010) through the Conflict in Adolescent Dating Relationships Inventory (CADRI, Wolfe et al., Reference Wolfe, Scott, Wekerle and Pittman2001) with sample of adolescent students: 96.3% of the sample indicated that they resorted to verbal violence and 24.3% to physical violence, while 95.4% suffered from verbal victimization and 21.7% from physical victimization, at least on one occasion. Furthermore, Carrascosa et al. (Reference Carrascosa, Cava and Buelga2015), also using the CADRI with adolescents from educational centers, found a prevalence of 21.6% for the perpetration of physical violence (14.3% of occasional violence and 7.3% of frequent violence), with the percentage of girls involved in occasional violence being higher. Regarding the perpetration of emotional violence, the prevalence was 74.4% (64.4% of occasional violence and 10% of frequent violence), with the percentage of girls involved in occasional violence being greater than that of boys.
There is no consensus regarding differences in DV according to sex. Thus, there are studies with general population that indicate that both members of the couple exhibit similar patterns of aggression (Fernández-González, O’Leary, & Muñoz-Rivas, Reference Fernández-González, O’Leary and Muñoz-Rivas2013). However, other studies such as Calvete, Gámez-Guadix, Fernández-González, Orue, and Borrajo’s (Reference Calvete, Gámez-Guadix, Fernández-Gonzalez, Orue and Borrajo2018) study, with a sample of Spanish adolescent students and using the Conflict in Adolescent Dating Relationships Inventory (CADRI, Wolfe et al., Reference Wolfe, Scott, Wekerle and Pittman2001), found that women showed higher levels of victimization than men. Jonson-Reid and Bivens’ (Reference Jonson-Reid and Bivens1999) study with adolescents in residential care is congruent with these results, finding more victimization in girls (37% vs. 11%), and slight differences in the perpetration of violence (16% girls vs. 15% Boys). On the contrary, Wincentak, Connolly, and Card’s (Reference Wincentak, Connolly and Card2017) recent meta-analysis did not find differences according to sex in physical victimization (21% for boys and girls) but found differences in the perpetration of violence (13% boys vs. 25% girls). In addition, Wekerle et al. (Reference Wekerle, Leung, Wall, MacMillan, Boyle, Trocme and Waechter2009) analyzed DV among adolescents being attended by Social Services using the CADRI and found percentages of 44% for perpetration of violence and 49% for victimization in boys in comparison to 67% of perpetration of violence and 63% of victimization in girls and noted that adolescents being attended by Child Protection Services are a group at special risk. The greater perpetration of physical and psychological violence on behalf of girls has also been corroborated by other studies (Fernández-Fuertes & Fuertes, Reference Fernández-Fuertes and Fuertes2010; Foshee et al., Reference Foshee, Reyes, Ennett, Suchindran, Mathias, Karriker-Jaffe and Benefield2011; Muñoz-Rivas et al. al., 2007; Pazos et al., Reference Pazos, Oliva and Hernando2014). However, as has been proven in the case of adults (Archer, Reference Archer2000), although the prevalence in physical and psychological perpetration on behalf of women is greater, this difference is minimal, and yet, the severity of the males’ behaviors is usually higher (Foshee et al., Reference Foshee, Benefield, Suchindran, Ennett, Bauman, Karriker-Jaffe and Mathias2009).
In relation to age, there are studies that indicate that, at a younger age, there is a higher risk of victimization in dating for girls (Bonomi et al., Reference Bonomi, Anderson, Nemeth, Bartle-Haring, Buettner and Schipper2012). Others point out that, with increasing age, violent behavior decreases (Redondo Pacheco, Inglés Saura, & García Lizarazo, Reference Redondo Pacheco, Inglés Saura and García Lizarazo2017). In this regard, the results derived from longitudinal studies should be highlighted, which, although scarcer, provide interesting results. Thus, Fernández-González, Calvete, and Orue (Reference Fernández-González, Calvete and Orue2017) found that DV (both victimization and perpetration) was quite stable over time (longitudinal study of 4 years with adolescents aged between 15 and 18 year), with psychological violence being greater than physical violence. The authors explained this greater stability of psychological violence due to its higher prevalence, and the adolescents’ perception of said violence as less antisocial and with lesser consequences. Other studies have corroborated the variability of physical violence over time, finding a curvilinear tendency. Thus, Foshee et al. (Reference Foshee, Benefield, Suchindran, Ennett, Bauman, Karriker-Jaffe and Mathias2009) analyzed the trajectory of adolescents from 13 to 19 years of age and found an upward trend from 13 to 16 or 17 years, and a decreasing trend from that age onwards. This decreasing tendency was also found in Nocentini, Menesini, and Pastorelli’s (Reference Nocentini, Menesini and Pastorelli2010) longitudinal study, especially in the case of girls. No study has been found that analyzes DV according to age in adolescents under residential care.
In relation to the individual factors related to DV, the fact that violence occurs in adolescence increases the severity of this phenomenon, due to the vulnerability of this developmental stage and the long-term consequences that this type of violence could have on the psychosocial development of these young people (Moreno, Estévez, Murgui, & Musitu, Reference Moreno, Estévez, Murgui and Musitu2009). During this developmental stage, self-esteem levels decrease, and depressive symptomatology increases, especially in girls, which may be an indicator of greater emotional distress and greater vulnerability (Díaz-Aguado, Reference Díaz-Aguado2003; Jaureguizar, Bernaras, Soroa, Sarasa, & Garaigordobil, Reference Jaureguizar, Bernaras, Soroa, Sarasa and Garaigordobil2015). In this regard, the results on DV indicate that both perpetration and victimization are related to anxiety and depression (Foshee et al., Reference Foshee, Reyes, Ennett, Suchindran, Mathias, Karriker-Jaffe and Benefield2011; Haynie et al., 2015; Holt & Espelage, Reference Holt and Espelage2005). In addition, victims of DV also have lower self-esteem and their self-concept is also affected (Carrascosa, Cava, Buelga, & Ortega, Reference Carrascosa, Cava, Buelga, Ortega and Castejón2016; Penado Abilleira, & Rodicio-García, Reference Penado Abilleira and Rodicio-García2017). Regarding the perpetration of violence, it has been found that people who resort to DV also have a lower self-concept (Carrascosa et al., Reference Carrascosa, Cava and Buelga2015).
On the other hand, several authors have highlighted the importance of social and cultural factors in the appearance of DV. In this sense, the sexist attitudes that support stereotypes and gender roles, both hostile and benevolent, play a central role in maintaining sex inequalities, and therefore, DV (Pazos et al., Reference Pazos, Oliva and Hernando2014). Soler, Barreto, and González (Reference Soler, Barreto and González2005) pointed out that both adolescent girls and adolescent boys, with more traditional ideas towards gender roles, accept the use of psychological, physical and sexual aggression towards the female sex more than towards the male sex. Other authors such as Karakurt and Cumbie (Reference Karakurt and Cumbie2012) found that sexist attitudes and values of egalitarianism in men were not associated with aggression towards their female partner or with the likelihood of being victimized by their partner. However, women with a lower level of hostile and benevolent sexism and with more egalitarian values were more likely to be attacked by their male partners.
As a result of the above and given the limited research on DV in adolescents under residential care, it is essential to study this type of violence to know its current situation and raise its visibility. In addition, in order to detect and prevent this phenomenon at an early age and, specifically, in adolescents under residential care, it is necessary to analyze the factors related to the perpetration of violence and victimization among young couples. In this way, violent relationships could be prevented from becoming chronic in adulthood and guaranteeing greater psychosocial well-being among minors.
The following objectives and hypotheses are proposed:
Objective 1: To analyze the prevalence of DV among adolescents under residential care based on sex and age.
Hypothesis 1: A prevalence for DV (perpetration and victimization) close to 48% (as that found by Jonson-Reid, et al., Reference Jonson-Reid, Scott, McMillen and Edmond2007), and a higher prevalence for verbal-emotional and relational rather than physical perpetration of violence and victimization are expected to be found.
Hypothesis 2: Greater victimization among women in the sample is expected to be found, and no differences in perpetration according to sex.
Hypothesis 3: It is expected that as age increases, the rate of DV will also increase.
Objective 2: To explore the relationship between victimization and the perpetration of DV, sexist attitudes and clinical variables.
Hypothesis 4: A positive relationship between the perpetration of violence, victimization and sexist attitudes is expected to be found.
Hypothesis 5: A positive relationship between DV (perpetration and victimization) and depression and anxiety is expected to be found, while the relationship of DV with self-esteem will be inverse.
Objective 3: To identify variables associated with the victimization and perpetration of DV in adolescents under residential care, using linear regression models.
Hypothesis 6: It is expected that anxiety, depression, self-esteem and sexism will be variables associated with DV (victimization and perpetration).
Method
Participants
The present study involved 271 minors under guardianship by the Provincial Councils of the Basque Country (Bizkaia, Gipuzkoa, and Alaba), aged between 12 and 17 years of age (M =15.23, SD =1.60), of which 54.6% (n = 148) were boys and 45.4% (n = 123) were girls. Regarding their origin, 54.6% (n = 148) came from the Basque Country, 29.5% (n = 80) were of foreign origin and 15.9% (n = 43) were from the rest of Spain.
The participants were housed in residential resources of different characteristics: Basic, specialized and emancipation programs. These resources are intended to respond to the multiple and diverse needs of teenagers, with their tutelage being granted to the Diputación (State). In relation to the residential resources of these minors, it should be noted that 65.7% (n = 177) of the adolescents in this study resided under the general basic program (the central and basic core of any foster care program that responds to the needs of children and adolescents between the ages of 4 and 18 years), 25.1% (n = 68) of adolescents in specialized programs for those with serious or very serious behavioral problems (aimed at adolescents between 13 and 18 years of age, presenting particularly disruptive behavior) and 9.6% (n = 26) in emancipation programs (residential equipment installed in ordinary homes that offer a service of assistance and education for adolescents over 16 years of age in order to facilitate their personal, social and work autonomy process).
In total, 69 residential resources were contacted out of the 83 centers with the described characteristics that currently exist in the Basque Country.
Instruments
Conflict in Adolescent Dating Relationships Inventory CADRI (Wolfe et al., Reference Wolfe, Scott, Wekerle and Pittman2001). The original scale consists of 70 items that assess both violent behavior towards the partner, as well as experiences of victimization. The present study was carried out using the version Carrascosa, Cava, and Buelga used in their study (2018). Specifically, the questionnaire consists of 17 items that analyze the different types of perpetration of violence; perpetration of relational violence (e.g., "I said things to his/her friends about him/her to make them go against him/her"), perpetration of verbal-emotional violence (e.g., "I brought up in conversation something bad that he/she had done in the past" ) and perpetration of physical violence (e.g., "I pushed him/her or I shook him/her"). In addition, another 17 items measure victimization: Relational victimization (e.g., "He/she tried to separate me from my group of friends"), verbal-emotional victimization (e.g., "He/she insulted me with put downs") and physical victimization (e.g., "He/she threw something at me"). Adolescents are asked to identify how often they have experienced these situations in their romantic relationship: Never -this has not happened in our relationship-, rarely -1 or 2 occasions-, sometimes - between 3 and 5 times - or frequently -6 or more occasions-. In the present study, the reliability coefficient, Cronbach’s alpha, for the total scale of violence perpetration was 0.72 and for the relational, verbal-emotional and physical violence perpetration subscales it was, respectively, 0.57, 0.79 and 0.77. The reliability of the total scale of victimization was 0.78 and for the relational, verbal-emotional and physical victimization subscales, it was 0.61, 0.86 and 0.83, respectively. The total alpha coefficient for this sample was .75.
Behavior Assessment System for Children and adolescents (BASC, Reynolds & Kamphaus, Reference Reynolds and Kamphaus1992). The S3 personality self-report is an inventory that consists of 185 statements with dichotomous answers (true/false), grouped into 14 scales, although for the present study, only the Anxiety, Depression and Self-esteem scales were collected. The internal consistency of these scales was Cronbach’s alpha = .81 for anxiety, Cronbach’s alpha = .81 for depression and Cronbach’s alpha = .84 for self-esteem.
Ambivalent Sexism Inventory for adolescents (ASI-A, Glick & Fiske, Reference Glick and Fiske1996). It is a 20-item scale with Likert-type responses of 6 options, ranging from 1 = Strongly disagree to 6 = Strongly agree. The ASI-Adolescents is an adaptation of the ASI (Inventory of Ambivalent Sexism by Glick & Fiske, Reference Glick and Fiske1996), for the adolescent population, which provides a measure of hostile sexism (HS) and another measure of benevolent sexism (BS). Hostile sexism refers to attitudes that are based on the inferiority of girls with respect to boys. It is assessed through 10 items such as: "Girls are easily offended", "Boys must control their girlfriends’ friendships". Benevolent sexism refers to a whole set of sexist attitudes towards girls where they are considered in a stereotyped way and limited to certain roles. This dimension also consists of 10 items, such as "Girls should be loved and protected by boys", "Relationships are essential to achieve true happiness in life". In the sample studied, the reliability index of the scale was Cronbach’s alpha = .80 (hostile sexism α = .67 and benevolent sexism α = .77).
Procedure
A first contact was made with the Protection of Children in Residential Care Section of the provinces of the Basque Country, which gave their approval for the investigation to take place. Subsequently, a commitment was signed and the data regarding the resources and those responsible for the study were submitted and approved by the Ethics Committee of the UPV/EHU [M10/2016/158].
A first contact was made with those responsible for the 83 residential resources (basic, specialized and emancipation programs), of which a total of 69 finally chose to participate in the present study. Appointments were made for the administration of the battery of tests, which were applied in paper format in person in a collective and individual manner. For the collection of data, all the regulations established by Organic Law 15/99 on the Protection of Personal Data were followed. In addition, the participants were informed of the voluntary nature of their participation and their necessary commitment in order to start administering the tests.
The minors without guardianship by the Provincial Councils of the Basque Country were excluded from the sample. In addition, all minors with mental health pathologies, intellectual disabilities or with a lack of language comprehension (for UFMs, questionnaires were collected only from adolescents living in the Basque Country for more than 4 years and with a good understanding of the language of the host country) were excluded. Those adolescents who reported not having a current partner or not having had a romantic partner in the past 12 months at the time of data collection were also excluded. Similarly, those participants whose answers were not reliable according to the validity indexes of the BASC–S3 test were removed from the study and their data destroyed. In total, 33 cases were excluded, 5 of them for never having had a partner, 23 for not exceeding 4 years in the host country, 3 with a diagnosis of mental health illness or intellectual disability and 2 whose tests were not properly completed and did not comply with the validity indexes of the BASC–S3 test.
Data analyses
The data analyses were carried out using the statistical package SPSS V.24. Descriptive analyses were performed in order to study the frequency of perpetration of violence and victimization. Following the criteria from previous studies (see Buelga, Iranzo, Cava, & Torralba, Reference Buelga, Iranzo, Cava and Torralba2015), those adolescents with scores above the mean plus a standard deviation (scores above 15) were assigned to the group of "frequent violence" and the adolescents with scores below the mean minus a standard deviation (scores greater than 0 but equal to or less than 15) were assigned to the "occasional violence" group. The same procedure was carried out with the victimization data, in this case considering the cut-off point of 16.9.
On the other hand, 2 (sex: boy vs. girl) x 2 (12–14 years vs. 15–17 years) analyses of variances were carried out to analyze the differences in victimization and perpetration of violence according to sex and age. Correlations were calculated between the perpetration of violence and victimization in relation to sexist attitudes and clinical variables (anxiety, depression and self-esteem). Finally, multiple linear regression analyses were carried out to identify variables associated to victimization and DV perpetration.
Results
Perpetration of violence and victimization according to sex and age
A total of 91.5% of adolescents reported having been violent in their relationships (perpetration), at least once in the past year and 88.6% said they had lived through violent situations (victimization). Regarding the frequency of DV perpetration, 77.5% (n = 210) reported being violent occasionally and 14% (n = 38) frequently. As far as victimization is concerned, 74.5% (n = 202) indicated that they had been an occasional victim, and 14% (n = 38) frequently suffered victimization in their dating relationships. Taking into account the different types of DV perpetration, 39.1% (n = 106) of the participants reported resorting to the perpetration of relational violence in their relationships, 89.3% (n = 242) reported perpetrating verbal-emotional violence and 38.7% (n = 105) perpetrating physical violence. Regarding victimization, 48.7% (n = 132) indicated suffering relational victimization, 87.1% (n = 236) reported suffering verbal-emotional victimization and 30.3% (n = 82) suffered physical victimization.
In addition, the possible differences in the total perpetration of violence and victimization according to sex and age were analyzed. No differences were found according to sex, F(1, 267) = .058, p = .001, ɳ2 = .013, yet differences were found according to age, F(1, 267) = 23.83, p = .001, ɳ2 = .082, with the means of adolescents aged between 12 and 14 years (M = 5.07, SD = 6.57) being lower than those of adolescents aged between 15 and 17 (M = 5.07, SD = 6.57) (see Figure 1).
Figure 1. Perpetration of Violence in Dating Relationships based on Sex and Age.
In relation to the victimization suffered, a significant interaction between sex and age was found, F(1, 267) = 5.20, p =. 023, ɳ2 = .019. As shown in Figure 2, the interaction shows that the difference between male and female victimization levels was greater in the 15–17 age group (women M = 12.89; SD = 10.06 and men M = 7.47; SD = 6.34) than in the 12–14 age group (women M = 5.81; SD = 4.94 and men M = 5.06; SD = 7.81).
Figure 2. Victimization in Dating Relationships based on Sex and Age.
Victimization and perpetration of DV and their relationship with sexist attitudes and clinical variables
As shown in Table 1, the perpetration of violence and victimization were positively and significantly related to hostile sexism, benevolent sexism, anxiety and depression. An inverse correlation was found between self-esteem and relational victimization (r = –.27) and verbal-emotional victimization (r = –.21).
Table 1. Correlation Matrix for the Studied Variables
Note. *p < .05. **p < .01.
Factors associated with the victimization and perpetration of DV
In order to find the factors that are associated with the victimization and perpetration of DV, multiple linear regressions were carried out on the total perpetration scale on the one hand and, on the scale of total victimization, on the other hand.
Table 2 shows that hostile sexism (β = .201, p = .032), age (β = .314, p = .001) and depression (β = .198, p = .041) were the variables significantly associated with the perpetration of DV.
Table 2. Variables Associated with the Total Perpetration of Violence of Minors under Residential Care
Note. R 2 = .20. p < .001.
β = beta the probability of making a Type 2 error in hypothesis
Table 3 shows that hostile sexism (β = .180, p = .008), depression (β = .138, p = .047), age (β = .207, p = .001) and sex (β = 241, p = .001) were the variables significantly associated with victimization.
Table 3. Variables Associated with the Total Victimization of Minors under Residential Care
Note. R 2 = .17, p <.001.
β = beta the probability of making a Type 2 error in hypothesis
Discussion
The present study, performed with a sample of 271 adolescents under residential care in the Basque Country, in contrast to previous studies on DV in schools or universities, contributes to the scientific community as one of the few studies with an adolescent sample of minors under the care of the Protection Services.
With respect to the first objective and the first hypothesis of this study, it could be indicated that the prevalence of DV found in the present study is higher than that found in the two previous studies with adolescents under residential care (Jonson-Reid & Bivens, Reference Jonson-Reid and Bivens1999; Jonson-Reid et al., Reference Jonson-Reid, Scott, McMillen and Edmond2007). However, it should also be noted that the studies are difficult to compare, as, on the one hand, the two previous studies did not consider the frequency of violence (occasional or frequent), and they were carried out with very specific samples. The sample in Jonson-Reid and Bivens’ (Reference Jonson-Reid and Bivens1999) study, in addition to being small, was composed of adolescents who had previously received an informative talk about DV, which could bias their answers; on the other hand, the sample in Jonson-Reid et al.’s (Reference Jonson-Reid, Scott, McMillen and Edmond2007) study was composed of 17-year-old adolescents (yet the sample of the present study was composed of adolescents with a wider age range: Between 12 and 17 years of age).
Comparing the present study with other studies carried out with adolescents from schools, the prevalence found in the present study is higher than that of previous international studies (Haynie et al., Reference Haynie, Farhat, Brooks-Russell, Wang, Barbieri and Iannotti2013; Vagi et al., Reference Vagi, Olsen, Basile and Vivolo-Kantor2015), although Holt and Espelage’s (Reference Holt and Espelage2005) study reported even higher figures of physical victimization than those of the present study. If the present study is compared with other studies carried out with Spanish samples, and with the same instrument (CADRI) (such as that of Cava et al., 2015; and that of Fernández-Fuertes & Fuertes, Reference Fernández-Fuertes and Fuertes2010), it is observed that adolescents under residential care show higher percentages in perpetration of physical violence, although the perpetration percentages of emotional-verbal violence are similar or even lower (although they are still very high). In general, and as expected (Hypothesis 1), a higher prevalence is found for the verbal-emotional and relational perpetration of violence and victimization than for physical violence and victimization.
Regarding the differences according to sex and age, as expected (Hypothesis 2) and in line with what was pointed out by Jonson-Reid and Bivens (Reference Jonson-Reid and Bivens1999) in their sample with adolescents under residential care, there are no differences in perpetration between boys and girls, but differences are found in victimization (higher in girls). These results are in line with those found in Calvete et al.’s (Reference Calvete, Gámez-Guadix, Fernández-Gonzalez, Orue and Borrajo2018) DV victimization study with adolescents from educational centers. However, it is essential to note the importance of taking into account not only sex, but also age, and both variables together, when analyzing the results, as the results clearly show that, as expected (Hypothesis 3), the greater the age, the higher the DV rate (perpetration and victimization), and that, in the specific case of victimization, in girls, this growing tendency is more evident. This may be due to the changes that take place in dating relationships, understanding that in the earliest periods of adolescence, such as 12 and 14 years old, there are usually no stable dating relationships across time, but among the 15- and 17-year-olds, dating relationships tend to be more durable.
Regarding the second of the proposed objectives, the results show that there is a positive relationship between the victimization and perpetration of DV and sexist attitudes, confirming Hypothesis 4. In addition, victimization is positively correlated with anxiety and depression, in the line of previous studies (Holt & Espelage, Reference Holt and Espelage2005), and negatively with self-esteem, coinciding with what has been found in the general adolescent population (Carrascosa et al., Reference Carrascosa, Cava, Buelga, Ortega and Castejón2016; Penado Abilleira & Rodicio-García, Reference Penado Abilleira and Rodicio-García2017), confirming Hypothesis 5.
Thus, in this study, it has also been found that social and cultural factors have an important weight in terms of DV. The result found between sexism and the victimization and perpetration of violence is in line with other studies that mention sexism among others as a determinant of violence (Pazos et al., Reference Pazos, Oliva and Hernando2014). Maintaining these sexist attitudes fosters the support of stereotypes and gender roles that disfigure the reality of relationships, which in turn can affect the psychosocial development of adolescent couples, due to the developmental stage in which they find themselves. Therefore, it is of vital importance to emphasize this type of violence, since many of the patterns in dating relationships that are observed in adolescence, can be repeated in adulthood.
Finally, it was hypothesized that anxiety, depression, self-esteem and sexism would be variables associated with DV (victimization and perpetration), which has been partially confirmed. The results show that age, hostile sexism and depression are variables associated with DV perpetration. Regarding victimization, the identified risk factors have been age, sex, depression and hostile sexism. These results emphasize the importance of taking into account and taking special care of the psychological well-being of minors. The presence of depressive symptoms associated with victimization and perpetration of violence in dating relationships may be showing the emotional distress of these adolescents, and their difficulties in dealing with their problems. In addition, there are studies that point out that minors with long stays under residential care have more emotional problems such as, for example, depression (Reference Bravo and Del ValleBravo & Del Valle, 2009).
On the other hand, the fact that sexist attitudes (specifically hostile sexism) are identified as associated variables in both regression models highlights the need to carry out intervention programs aimed at young adolescents, with the objective of modifying those sexist behavior patterns internalized from an early age. Also, it would be necessary to carry out programs to prevent sexist attitudes aimed at minors under a residential care regime. Therefore, the Basic Social Services should consider carrying out programs for the prevention of violence at an early age, since it is in these services that the first contact with these minors is established.
Finally, it is worth noting the need to analyze DV from a developmental point of view. The fact that age is another important factor to consider indicates that we must take into account the developmental changes that occur in these ages. More longitudinal studies should be considered, which would provide more information about the changes that occur throughout the adolescent years.
Regarding the limitations of this study, the design of the study (transversal) and that the order of administration of the instruments was not counterbalanced should be mentioned. These limitations may have had some impact on the results obtained.
