Witnessing Partner Violence

Abstract

Research has shown that witnessing partner violence (WPV) increases the likelihood of experiencing or perpetrating violence in later romantic relationships, but little is known about the mechanisms underlying this process. This study examines the relationships between preference for unsuitable partners and teen dating violence (TDV) among adolescents who have witnessed parental violence or not. Attachment was also considered. Participants were 356 adolescents, both witnesses and non-witnesses of partner violence. Results showed no difference in preferences (for good, risky, or loving partners) between the two groups. However, preference for unsuitable partners did significantly predict TDV perpetration and victimization, but only among witnesses. Also, loving-partner preference moderates the relationship between WPV and TDV perpetration among highly avoidant witnesses. Findings indicate a new avenue for prevention through targeting partner preferences.

Keywords

teen dating violence witnessing partner violence partner preferences attachment adolescents

Research has shown that adults and adolescents who were exposed to intimate partner violence between their parents have an increased likelihood of perpetrating or experiencing violence in their own intimate relationships (Kaukinen, 2014; Temched et al., 2008; Vani et al., 2013), along with suffering other negative consequences (Kitzmann, Gaylord, Holt, & Kenny, 2003). This association was initially attributed to social learning, giving rise to the hypothesis of intergenerational transmission of family violence (Straus, Gelles, & Steinmetz, 1980). However, the process of witnessing partner violence (WPV) has proved to be more complex than first thought. For example, only about 30% of men in the community who witnessed this type of violence during their childhood become violent with their intimate partners (DelSol & Margolin, 2004). After abandoning the idea that children are passive recipients of what they observe, researchers have begun to examine the role of a number of factors that might explain differences detected in this process (Godbout, Dutton, Lussier, & Sabourin, 2009; Kinsfogel & Grych, 2004; Tajina, Herrenkohl, Moylan, & Derr, 2010).

Attachment styles are among the factors being analyzed to better understand the consequences of exposure to parental violence. According to attachment theorists, the quality of the child−parent attachment bond affects social and emotional development, through rudimentary conceptual representations of the self and others (Bartholomew & Horowitz, 1991; Bowlby, 1973). It is assumed that early experiences with caregivers lead to different attachment styles, which emerge from two underlying dimensions: anxiety over abandonment and avoidance of intimacy. These attachment styles continue to guide the individual’s emotions, cognitions, and behaviors in later relationships (Simpson & Rholes, 2010). In this sense, evidence indicates that the quality of parenting is clearly affected when parental violence occurs, and that children tend to respond with a highly negative affect and cognition (Lloyd, 2013). Furthermore, research has found that early experiences of violence relate to adult intimate partner violence, both directly and indirectly, through insecure attachment (Godbout et al., 2009). In addition, insecurely attached individuals (i.e., those who score higher in anxiety and/or avoidance) are more likely to be both victims and perpetrators of dating aggression in their relationships (Karakurt, Keiley, & Posada, 2013).

Relationship dynamics are predicted by attachment styles but may also be affected by the choice of romantic partners. Indeed, choosing unsuitable partners may increase the likelihood of conflicts and violence in relationships. There is as yet little evidence linking attachment styles, partner preferences, and dating violence. However, some findings point in this direction. Although Cohen and Belsky (2008) found that female mate preferences are condition-dependent, they also detected that female dismissing-avoidant respondents showed different interests (e.g., for shorter-term relationships) than preoccupied or secure respondents. In a similar vein, Shorey, Cornelious, and Bell (2008) suggested that insecurely attached individuals seem to gravitate toward partners who may be abusive. Moreover, Zayas and Shoda (2007) have noted that intimate partner choices might facilitate the development of violent relationships, because the preferences shown by male perpetrators and female victims seem to lead them to choose each other. On the assumption that a relationship between these factors does exist, it is possible to take it a step further and hypothesize a connection with WPV. Thus, for individuals who had been exposed to parental violence, there would be an unfavorable interplay between disposition and environmental conditions if they chose unsuitable partners. By contrast, if they chose more suitable partners, they would face a lower risk in their intimate relationships.

According to the General Aggression Model (GAM), both personal and situational factors significantly increase the likelihood of different forms of violence, including that directed against an intimate partner (DeWall, Anderson, & Bushman, 2011). In this sense, WPV, insecure attachment, and unsuitable preferences may be conceptualized as personal characteristics that predispose one to TDV. However, although research has proved the link between violence and the first two factors, the association between these constructs and unsuitable preferences has not yet been analyzed. Moreover, dysfunctional dynamics have been signaled among those situational factors responsible for eliciting TDV, which is supposed to be more likely among those who are predisposed to responding aggressively. From this framework (Figure 1), this study is aimed at examining the relationships between dating partner preferences, attachment styles, and TDV among adolescents who report either having witnessed partner violence between their parents or not. Specifically, the hypotheses are as follows:

Hypothesis 1: The preference for dating partners with unsuitable characteristics will be greater among those participants who report WPV involving their parents (witnesses) than among those who do not (non-witnesses).

Hypothesis 2: The greater the preference for dating partners with unsuitable characteristics, the higher the level of TDV perpetration and victimization, but only among those who report WPV involving their parents (witnesses).

Hypothesis 3: The greater the exposure to intimate partner violence involving parents and the preference for unsuitable partners, the higher the level of TDV perpetration and victimization, but only among the insecurely attached (high anxiety and/or high avoidance).

Figure 1.

Theoretical model describing proposed link between WPV, insecure attachment, unsuitable partner preferences, and TDV.

Method

Participants and Procedure

A subsample of a large sample of adolescents was selected for this study. Participants were 356 adolescents (190 girls and 166 boys) who currently had—or had had in the past—at least one opposite-sex partner (n = 322), same-sex partner (n = 7), or both types of partners (n = 27). Their ages ranged from 12 to 20 (M = 15.66, SD = 1.23 years). Among participants, 19.9% claimed to have witnessed some degree of parental violence (witnesses).

Before the start of the study, permission was requested from participants’ high schools and families. In addition, compliance with ethical standards has been positively assessed by the Institutional Review Board. Data collection was carried out by means of a questionnaire given to all students during school hours. All participants received identical instructions and responded to the same questionnaire. Student participation was voluntary, and participants were assured in advance of their anonymity. No participants received extra credit, but only 0.5% refused to participate.

Measures

Along with questions about demographic characteristics and romantic experience, the instrument consisted of several scales.

Attachment

Attachment styles were measured through the Spanish adaptation (Fernández-Fuertes, Orgaz, Fuertes, & Carcedo, 2011) of the Experiences in Close Relationships−Revised Scale (ECR-R; Fraley, Waller, & Brennan, 2000). This 18-item scale ranges from 1 (strongly disagree) to 7 (strongly agree), with 9 items to assess anxiety about abandonment and another 9 items to measure avoidance of intimacy. Scores on each of these dimensions allow for the description of different attachment styles. Thus, individuals who score lower in both dimensions are considered as secure, those who score higher in avoidance and lower in anxiety are described as dismissing-avoidant, those who score higher in anxiety and lower in avoidance are preoccupied, and those who score higher in both dimensions are fearful. Internal consistency was obtained through Cronbach’s alpha, reaching values of .87 and .86, respectively.

TDV victimization/perpetration

Psychological abuse and physical violence were measured through two different scales. First, the Safe Dates–Psychological Abuse Victimization subscale, developed by Foshee et al. (1998), was used for assessing both psychological abuse victimization and perpetration. This subscale consists of 14 items that measure the following: verbal aggression (said things to hurt my feelings on purpose, brought up something from the past to hurt me, etc.), control of intimate partner (told me I could not talk to someone of the opposite sex, etc.), and interrupted physical aggression (threw something at me but missed, etc.).

Second, a shortened version of the Conflict Tactics Scale (CTS; Straus, 1979) was used to measure physical violence victimization and perpetration. In all cases, participants were required to estimate both how often different behaviors were shown by themselves (perpetration) or their partners (victimization). Responses for both scales ranged from 0 (never) to 3 (very often). Finally, two global TDV indexes were computed (one for perpetration and another for victimization), adding all items of the Psychological Abuse scale and two of the physical violence items (pushing and hitting). Cronbach’s alpha reached a value of .82 for both perpetration and victimization.

WPV

The same version of the CTS (Straus, 1979) was used to measure WPV involving their parents. In this case, participants were required to estimate how often they had witnessed their fathers and mothers showing these behaviors during their conflicts. Their responses ranged from 0 (never) to 10 (very often). Four items of this scale (threatening with harm, throwing objects, pushing, and hitting) were used to compute a global parental violence index (α = .90), adding fathers’ and mothers’ scores.

Dating partner preferences

After reviewing the literature on intimate partner preferences (Lippa, 2007; Zayas & Shoda, 2007), we carried out an initial study analyzing 34 desirable and risk partner characteristics (Gonzalez-Mendez, Martín, & Hernández-Abrante, 2014). For the present study, three scales were selected to assess preference for three types of dating partners: good, risky, and loving. The first comprised four traits (intelligent, honest, kind, and with a sense of humor), the second had five (loves risk, likes to break the rules, rebellious, jealous, and controlling), and the third included two traits (romantic and affectionate). These three scales were used to measure participants’ dating partner preferences. Responses ranged from 0 (not at all) to 10 (very much). Cronbach’s alphas for these scales ranged from .65 to .72.

Results

To test Hypotheses 1 and 2, we first computed a composite index including any aggression from either parent toward the intimate partner. Subsequently, participants were classified into two groups. Those who reported having witnessed any form of intimate partner violence involving parents (frequency > 0) were categorized as witnesses, and those who did not (frequency = 0) were categorized as non-witnesses. Zero-order correlations for each of these groups are shown in Table 1.

Table 1.

Correlations, Means, and Standard Deviation Among Variables.

	Non Witnesses (n = 144)			Witnesses (n = 114)			1	2	3	4	5	6	7
	M	SD	Range	M	SD	Range	1	2	3	4	5	6	7
1. TDV perpetration	0.18	0.20	0-1	0.26	0.36	0-2.56		.76**	−.09	.06	−.14	.10	.10
2. TDV victimization	0.23	0.28	0-1.81	0.32	0.37	0-1.81	.76**		−.11	.20	−.22**	.20*	.11
3. Preference for good partners	8.24	1.05	4.50-10	8.29	1.18	3-10	−.10	−.09		−.01	.21**	−.32**	.01
4. Preference for risk partners	3.73	1.55	0-8.60	3.77	1.70	0-10	.16	.09	−.02		.15*	−.05	.22**
5. Preference for loving partners	8.40	1.39	3-10	8.44	1.67	2-10	.07	.00	.18*	.03		−.29**	.24**
6. Avoidance	2.47	1.08	1-6.44	2.33	1.07	1-5.44	−.02	.13	−.21*	−.07	−.32**		.04
7. Anxiety	3.65	1.43	1.11-6.78	3.74	1.37	1-6.67	.31**	.31**	−.10	.15	.13	.05

Note. Correlations for non-witnesses are reported above the diagonal, and those for witnesses below the diagonal. TDV = teen dating violence.

p < .05. **p < .01.

Levene’s test indicated that the variances of the two populations were not homogeneous with respect to TDV perpetration and victimization. Thus, Welch’s test was carried out to compare witness and non-witness participants on these measures. Results confirmed significant differences for both TDV perpetration F(1, 206.179) = 6.07, p < .05, η_p² = .02, and victimization, F(1, 246.583) = 6.06, p < .05, η_p² = .02. Specifically, witnesses reported a higher level of TDV perpetration (M = 0.26, SD = 0.36) and victimization (M = 0.32, SD = 0.37) than non-witnesses (M = 0.18, SD = 0.20 and M = 0.22, SD = 0.28, respectively). By contrast, an ANOVA did not show any significant differences regarding preferences, which allowed us to reject Hypothesis 1.

To test Hypothesis 2, the data analyses followed several steps. First, the 33rd and 66th percentile composite scores on TDV perpetration and victimization were determined separately. Participants who scored below the 33rd percentile, in each of these two measures, were classified as “low” and those who scored larger than the 66th percentile were classified as “high”. Then, these participants were selected for subsequent analyses. In a second step, witness and non-witness participants were separated before binary logistic regression analyses were performed to predict either TDV perpetration or victimization in each of these two groups.

As stated in Hypothesis 2, a preference for unsuitable partners should significantly predict TDV perpetration among witnesses but not among non-witnesses. Specifically, belonging to the most violent group was predicted by a higher preference for risky partners (B = 0.46, Wald’s test = 4.13, df = 1, p < .05, SE = .23, odds ratio [OR] = 1.59, 95% confidence interval [CI] = [1.02, 2.49]) and a lower preference for good partners (B = −0.54, Wald’s test = 4.30, df = 1, p < .05, SE = .26, OR = 0.58, 95% CI = [0.35, 0.97]). This model allows for the correct classification of 62.6% of witnesses (68.8% of the true-negatives and 55.8% of the true-positives).

In the same way, a lower preference for good partners increases the likelihood of TDV victimization among witnesses but not among non-witnesses (B = −0.52, Wald’s test = 4.74, df = 1, p < .05, SE = .24, OR = 0.59, 95% CI = [0.37, 0.95]). In this case, the model correctly classified 63.5% of the witnesses (55.8% of the true-negatives and 69.8% of the true-positives).

Regarding attachment, an ANOVA was carried out to compare witnesses’ and non-witnesses’ scores on anxiety and avoidance. Results did not show significant differences in any case. Several steps were then followed to test Hypothesis 3. First, the 33rd and 66th percentile composite scores on anxiety and avoidance dimensions were determined separately, and participants were classified as “low” (those who scored below the 33rd percentile) and as “high” (those who scored larger than the 66th percentile) in each of these two attachment styles. Second, binary logistic regression analyses to predict TDV were performed after separating participants by their level of anxiety (high or low) and then by their level of avoidance (high or low). These regressions included the level of WPV and partner preferences as covariates.

When teens were separated according to their level of avoidance, a significant interaction between WPV and loving-partner preference predicted TDV perpetration among highly avoidant participants only (B = 1.97, Wald’s test = 4.45, df = 1, p < .05, SE = .94, OR = 7.21, 95% CI = [1.15, 45.13]). This model correctly classified 78.1% of the witnesses (90% of the true-negatives and 58.3% of the true-positives). As shown in Figure 2, those who are highly avoidant and less interested in loving partners show similar levels of TDV perpetration, irrespective of whether they are high or low in WPV. By contrast, TDV perpetration levels in teens who are highly interested in loving partners are higher when they are witnesses and lower when they are non-witnesses.

Figure 2.

TDV preparation among highly avoidant teens according to their level of WPV and preference for loving partners.

Discussion

Although research has shown that WPV increases the likelihood of experiencing and/or perpetrating violence in later romantic relationships (e.g., Temched et al., 2008), little is known about the mechanisms underlying this process. This study explores the role of partner preferences in the association between WPV and TDV perpetration/victimization, analyzing whether witnesses and non-witnesses show different preferences, and whether this predicts TDV perpetration and victimization.

As far as we know, research has not analyzed the relationship between WPV and partner preferences. However, several findings suggest that these are pieces of the same puzzle. As is the case with attachment, choosing unsuitable partners may facilitate dysfunctional dynamics and contribute to the development of violent relationships. In this sense, some researchers have suggested that insecurely attached individuals may gravitate toward partners who make abusive relationships more likely (Shorey et al., 2008) and that partner preferences may lead members of violent couples to choose each other (Zayas & Shoda, 2007). Moreover, research has also demonstrated the relationships between WPV, insecure attachment, and TDV perpetration/victimization (Godbout et al., 2009; Karakurt et al., 2013).

Taking all this into account, it was hypothesized that unsuitable partner preferences would be higher among teens who have witnessed intimate partner violence between their parents than among non-witnesses (Hypothesis 1), and that these preferences would be associated with a higher risk of TDV perpetration and victimization (Hypothesis 2). Contrary to the assertion in Hypothesis 1, our results showed no difference in partner preferences between witnesses and non-witnesses. However, although WPV does not consistently lead to unsuitable preferences, it is precisely those individuals who combine WPV and an interest for unsuitable partners who have a higher risk of TDV perpetration and victimization. Indeed, our results support Hypothesis 2, as unsuitable preferences only increase the likelihood of TDV among witnesses. Specifically, a lower interest for good partners (intelligent, honest, kind, and with a sense of humor) increases the likelihood of both experiencing and perpetrating TDV. In addition, the higher the interest in risk traits (loves risk, likes to break the rules, rebellious, jealous, and controlling), the higher the risk of TDV perpetration.

Different reasons may lead adolescents to prefer unsuitable partners. Among some peer groups, preference for certain partner characteristics such as rebelliousness or jealousy may be normative. However, the results of this study indicate that the risk that this entails is significantly greater in witnesses of partner violence than in non-witnesses. At least in part, it may be that unsuitable partners are more likely to contribute to dysfunctional dynamics, which will be more dangerous for those teens with more difficulties in regulating their emotions. In this way, choosing unsuitable partners would make teens who have witnessed partner violence more likely become involved in violent relationships.

Moreover, similarity is a well-documented predictor of attraction, and rebellious individuals and risk-takers tend to associate with others who are similar to them (Rhule-Louie & McMahon, 2007). Therefore, it is possible that those who prefer risky and not so good partners are more predisposed to problem behavior. Research has also demonstrated that WPV influences the acquisition of beliefs that violence is justifiable and effective, which increases the likelihood of engaging in aggression (Kinsfogel & Grych, 2004). Thus, teen witnesses of partner violence interested in unsuitable partners might also be more inclined to justify violence and become involved in episodes of aggression.

Hypothesis 3 was only confirmed among highly avoidant teens. In this case, a significant interaction between WPV and loving-partner preference supports the moderating role of preferences on transmission of the parental violence witnessed. Although witnesses and non-witnesses who expressed less interest in loving partners did not differ in TDV perpetration, those who expressed more interest in loving partners did show significant differences. Specifically, the highest level of TDV perpetration was shown by those highly exposed to WPV who prefer loving partners.

Although it seems reasonable that a preference for loving partners would be a protective factor among teens, the attachment perspective warns against this type of romantic longing. Among highly anxious people, a strong pursuit of emotional closeness usually associates with a poor self-image, which hinders close relationships (Bartholomew & Horowitz, 1991). Among highly avoidant teens, it might indicate a contradictory desire that often characterizes fearful-avoidant individuals. Therefore, those teen witnesses who show a high preference for loving partners would seem to pursue a difficult goal, that is, an affectionate partner who does not demand intimacy.

In summary, the results obtained support the need to analyze partner preferences to better understand the consequences of WPV on dating relationships. Unsuitable preferences are not inevitable among witnesses, but those who do show these preferences are at an increased risk of reproducing/experiencing such violence in their relationships. The findings also indicate a new avenue for prevention, because teens’ partner preferences may be targeted. However, this approach requires that we understand the functions that these preferences satisfy. Whether unsuitable preferences are normative in a peer group and/or whether they fit teens’ views of themselves and their relationships will result in different scenarios and will probably require different intervention strategies.

The conclusions discussed above should be considered as preliminary until the findings can be replicated in a larger sample of adolescents. Also, it is necessary to highlight some of the limitations of this study. First, WPV is frequently associated with several forms of maltreatment and exposure to other forms of family violence (Hamby, Finkelhor, Turner, & Ormrod, 2010). This means that the consequences of WPV may be different depending on whether it is accompanied or not by other forms of violence.

Moreover, there are some relevant variables from the peer context that have proved to be relevant in predicting TDV victimization and perpetration among those who have witnessed parental violence. Indeed, the evidence indicates that peer influence may be greater than family influence for dating violence (Arriaga & Foshee, 2004). In this sense, future research should pay more attention to peer influence on partner preferences, because shared romantic ideals for unsuitable partners may increase risk among those exposed to WPV.

Finally, it would also be desirable to analyze longitudinally the relationship between preferences and TDV among adolescents who either have witnessed intimate partner violence between their parents or not. In this sense, gender should also be taken into account in future analyses.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded in part by the Ministerio de Sanidad, Servicios Sociales e Igualdad (Spain) and European Social Fund under Grant 53/12.

Author Biographies

Rosaura Gonzalez-Mendez, PhD, is an associate professor in the Department of Cognitive and Social Psychology at the University of La Laguna, Spain. Her research interests include the etiology and prevention of dating violence.

José M. Yanes, MSc, is a doctoral candidate in the Clinical Psychology Program at the University of La Laguna.

Gustavo Ramírez-Santana, PhD, is an associate professor in methodology of the behavioral sciences at the University of La Laguna, Spain. He lectures on data analysis, and his research interests are related to applied studies and simulations about the robustness and power of different statistical techniques.

References

Arriaga

X. B.

Foshee

V. A.

(2004). Adolescent dating violence: Do adolescents follow in their friends’, or their parents’, footsteps? Journal of Interpersonal Violence, 19, 162-184. doi:10.1177/0886260503260247

Bartholomew

Horowitz

L. M.

(1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244. doi:10.1037//0022-3514.61.2.226

Bowlby

(1973). Attachment and Loss: Vol. 2. Separation: Anxiety and anger. London, England: Hogarth Press.

Cohen

D. L.

Belsky

(2008). Individual differences in female mate preferences as a function of attachment and hypothetical ecological conditions. Journal of Evolutionary Psychology, 6, 25-42. doi:10.1556/JEP.2008.1001

DelSol

Margolin

(2004). The role of family-origin violence in men’s marital violence perpetration. Clinical Psychology Review, 24, 99-122. doi:10.1016/j.cpr.2003.12.001

DeWall

C. N.

Anderson

C. A.

Bushman

B. J.

(2011). The general aggression model: Theoretical extensions to violence. Psychology of Violence, 1, 245-258. doi:10.1037/a0023842

Fernández-Fuertes

A. A.

Orgaz

Fuertes

Carcedos

(2011). La evaluación del apego romántico en adolescentes españoles: validación de la versión reducida del Experiences in Close Relationships-Revised (ECR-R) [Romantic attachment assessment in Spanish adolescents: Validation of the short version of the Experiences in Close Relationships-Revised (ECR-R)]. Anales de Psicología, 27, 827-833.

Foshee

V. A.

Bauman

K. E.

Arriaga

X. B.

Helms

R. W.

Koch

G. G.

Linder

G. F.

(1998). An evaluation of Safe Dates, an adolescent dating violence program. American Journal of Public Health, 88, 45-50. doi:10.2105/ajph.88.1.45

Fraley

R. C.

Waller

N. G.

Brennan

K. A.

(2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350-365. doi:10.1037//0022-3514.78.2.350

10.

Godbout

Dutton

D. G.

Lussier

Sabourin

(2009). Early exposure to violence, domestic violence, attachment representations, and marital adjustment. Personal Relationships, 16, 365-384. doi:10.1111/j.1475-6811.2009.01228.x

11.

Gonzalez-Mendez

Martín

Hernández-Abrante

(2014). At the end of a fairy tale. Romantic relationships in female juvenile offenders. Journal of Forensic Psychiatry and Psychology, 25, 584-599. doi:10.1080/14789949.2014.943794

12.

Hamby

Finkelhor

Turner

Ormrod

(2010). The overlap of witnessing partner violence with child maltreatment and other victimizations in a nationally representative survey of youth. Child Abuse & Neglect, 34, 734-741. doi:10.1016/j.chiabu.2010.03.001

13.

Karakurt

Keiley

Posada

(2013). Intimate relationship aggression in college couples: Family-of-origin violence, egalitarian attitude, attachment security. Journal of Family Violence, 28, 561-575. doi:10.1007/s10896-013-9526-9

14.

Kaukinen

(2014). Dating violence among college students: The risk and protective factors. Trauma, Violence, & Abuse, 15, 283-296. doi:10.1177/1524838014521321

15.

Kinsfogel

K. M.

Grych

J. H.

(2004). Interparental conflict and adolescent dating relationships: Integrating cognitive, emotional, and peer influences. Journal of Family Psychology, 18, 505-515. doi:10.1037/0893-3200.18.3.505

16.

Kitzmann

K. M.

Gaylord

N. K.

Holt

A. R.

Kenny

E. D.

(2003). Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology, 71, 339-352. doi:10.1037/0022-006x.71.2.339

17.

Lippa

R. A.

(2007). The preferred traits of mates in a cross-national study of heterosexual and homosexual men and women: An examination of biological and cultural influences. Archives of Sexual Behavior, 36, 193-208. doi:10.1007/s10508-006-9151-2

18.

Lloyd

S. A.

(2013). Family violence. In Peterson

G. W.

Busch

K. R.

(Eds.), Handbook of marriage and the family (pp. 449-485). Oxford, UK: Springer.

19.

Rhule-Louie

D. M.

McMahon

R. J.

(2007). Problem behavior and romantic relationships: Assortative mating, behavior contagion, and desistance. Clinical Child and Family Psychology Review, 10, 53-100. doi:10.1007/s10567-006-0016-y

20.

Shorey

R. C.

Cornelious

T. L.

Bell

K. M.

(2008). A critical review of theoretical frameworks for dating violence: Comparing the dating and marital fields. Aggression and Violent Behavior, 13, 185-194. doi:10.1016/j.avb.2008.03.003

21.

Simpson

J. A.

Rholes

W. S.

(2010). Attachment and relationships: Milestones and future directions. Journal of Social and Personal Relationships, 27, 173-180. doi:10.1177/0265407509360909

22.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The conflict tactics scale. Journal of Marriage and Family, 41, 75-88. doi:10.2307/351733

23.

Straus

M. A.

Gelles

R. J.

Steinmetz

S. K.

(1980). Behind closed doors: Violence in the American family. New York, NY: Anchor Books.

24.

Tajina

E. A.

Herrenkohl

T. I.

Moylan

C. A.

Derr

A. S.

(2010). Moderating the effects of childhood exposure to intimate partner violence: The roles of parenting characteristics and adolescent peer support. Journal of Research on Adolescence, 21, 376-394. doi:10.1111/j.1532-7795.2010.00676.x

25.

Temched

Serbin

Martin-Storey

Stack

Hodgins

Ledingham

Schwartzman

A. E.

(2008). Continuity and pathways from aggression in childhood to family violence in adulthood: A 30-year longitudinal study. Journal of Family Violence, 23, 231-242. doi:10.1007/s10896-007-9147-2

26.

Vani

K. J.

Rothman

Latzman

N. E.

Thar

A. T.

Hall

D. M.

Breiding

M. J.

(2013). Beyond correlates: A review of risk and protective factors for adolescent dating violence perpetration. Journal of Youth Adolescence, 42, 633-649. doi:10.1007/s10964-013-9907-7

27.

Zayas

Shoda

(2007). Predicting preferences for dating partners from past experiences of psychological abuse: Identifying the psychological ingredients of situations. Personality and Social Psychology Bulletin, 33, 123-138. doi:10.1177/0146167206293493