THE LONGITUDINAL IMPACT OF ABUSERSÕ USE OF CHILDREN ON IPV SURVIVORSÕ AND CHILDRENÕS WELL-BEING  By  Katie A. Gregory              A DISSERTATION  Submitted to Michigan State University in partial fulfillment of the requirements for the degree of PsychologyÑDoctor of Philosophy 2016       ABSTRACT THE LONGITUDINAL IMPACT OF ABUSERSÕ USE OF CHILDREN ON IPV SURVIVORSÕ AND CHILDRENÕS WELL-BEING By Katie A. Gregory  There is a well-established a connection between intimate partner violence (IPV) and its deleterious impact on survivors and their children. While there have been numerous studies examining IPV in terms of psychological, physical, and economic abuse, far fewer have looked at the abusersÕ use of children as a tactic of abuse and its impact on both survivorsÕ and their childrenÕs well-being. This study sought to add to the field by further exploring this relationship. The sample for this study, taken from a larger two-year longitudinal study examining the experiences of help-seeking IPV survivorsÕ and their children, includes 105 survivors across three time points over the course of eight months. Using a time-ordered mediation model, this study examined the relationship between the use of children at a first time point and its impact on survivorsÕ quality of life at eight months later as mediated by their childrenÕs behavior at a middle (four month) time point. Three models were examined in this study to understand the mediating effects of the childrenÕs behavior between the use of children and the survivorsÕ quality of life. The first model, a simple mediation model, included the three (independent, mediator, and dependent) variables. The second model was the same simple mediation model while controlling for the childrenÕs ages. The third and final model was the same as the second model  while also controlling for survivorsÕ quality of life at the first time point. The findings from this study are mixed, indicating a mediated relationship in models one and two, and no mediation in the third model (when controlling for quality of life at the first time point). Preliminary findings from this study indicate that when abusers use children in order to control their partners and ex-partners, the children display higher levels of negative behavior, which in turn negatively effects survivorsÕ quality of life yet the significance of these relationship may change when controlling for other variables.                                             Copyright by KATIE A. GREGORY 2016 v ACKNOWLEDGEMENTS  There are so many people who made graduate school and this dissertation possible. There were many times when it felt like this moment would never come. With the support, love, and encouragement from each of the following people it happened. To Justin, Eleanor, and Brandon. You have been my rocks. When my head got filled with ideas and I had a hard time letting go, I could always count on each of you to ground me the second we got home. Justin, your support and patience throughout my graduate career has been amazing. You have shown be what a true partnership can be like Ð by caring for our kids, making us countless meals, and letting me lean on you. From the bottom of my heart, thank you. Eleanor and Brandon, you continue to be my favorite surprises. You have brought so much joy and perspective to life and the work I do. I didnÕt realize how much I could love and laugh before you two came into my world.  Mom and Dad, you have come to expect me to forge my own path and I cannot express my gratitude enough for your support. From the time I challenged the status quo by auditioning to be Johnny Appleseed in the sixth grade, you have encouraged me to fight for what I thought was just and fair. Through your example I know that I have the ability to create change in the world and that even the most minor of actions can be the most meaningful -- even writing a dissertation. To my advisor and chair, Dr. Cris Sullivan, your guidance, mentorship and friendship have helped me grow in ways I never expected. You took a chance on me when I stepped back into your office after eight years of being away. I have the feeling neither of us regret that leap of faith today. Thank you.  vi To Dr. Rebecca Campbell, I will never forget one of our interactions during my first semester as a graduate student. I sat in your office with tears in my eyes and you told me that the faculty see me and know that I am putting forth the effort. You have repeatedly given me opportunities and reminded me of my abilities. Thank you. To Dr. Deb Bybee, thank you for your guidance over these last few rushed months. I sincerely thought my ability to fully understand quantitative research was almost a lost cause. You have sparked my confidence again. To Dr. Rub”n Parra-Cardona, your enthusiastic support has been key in my development as a scholar. You have set an amazing example as a scholar and a mentor and your standards have made my research stronger as a result. To all of my family and friends, this would never have happened without you. Jon, Heidi, and Emily, you have found ways to keep me laughing throughout the years. Your regular phone calls and check-ins have been more supportive than you realize. To the rest of the grandparents: GiGi, Grandpa Bill, Papa Dick, and Grandma Jill, thank you for stepping in when you could to help with the kids and support us as a family.  To my numerous friends made in graduate school, including Dr. Mariah Kornbluh, Dr. Adrienne Adams, Dr. Nidal Karim, Hannah, Dr. Nkiru Nnawulezi, Dr. Jess Shaw, Sara, Jenny, and Dani, thank you for becoming a part of my extended family through our countless hours of working side by side and sometimes together. To the women and children who were a part of this study, I am so grateful for you having shared your stories and experiences. Your participation years ago continue to have an impact. Finally, to all survivors of intimate partner violence: I am here. I see you. And I will do my best to honor you through my work in these difficult times.  vii TABLE OF CONTENTS    LIST OF TABLES .....................................................................................................................  ix   LIST OF FIGURES ....................................................................................................................  x   CHAPTER 1: INTRODUCTION ..............................................................................................  1 The Characteristics of Intimate Partner Violence ................................................................  2 Coercive Control ..............................................................................................................  2 Increased Lethality When Attempting to Leave the Abusive Partner .............................  3 Use of Children As a Form of IPV ..................................................................................  4 The Impact of Abuse on Survivors .......................................................................................  7 Physical Outcomes ...........................................................................................................  7 Psychological Outcomes ..................................................................................................  8 Quality of Life ..................................................................................................................  11 The Impact of IPV on Children ............................................................................................  13 Current Study .......................................................................................................................  15 Study Hypothesis .............................................................................................................  16   CHAPTER 2: METHOD ............................................................................................................  18 Procedures ............................................................................................................................  18 Recruitment ......................................................................................................................  18 Eligibility .........................................................................................................................  18 Interviews .........................................................................................................................  19 Participants ...........................................................................................................................  19 Demographics ..................................................................................................................  19 Survivors ....................................................................................................................  19 Target children ...........................................................................................................  20 Measures ...............................................................................................................................  23 AssailantÕs Relationship to the Survivor and Target Child .............................................  23 Intimate Partner Violence ................................................................................................  23 Physical abuse ............................................................................................................  23 Emotional abuse .........................................................................................................  24 Use of children ...........................................................................................................  24 ChildrenÕs Behavior .........................................................................................................  24 Quality of Life ..................................................................................................................  25 Intervention Condition .....................................................................................................  25   CHAPTER 3: ANALYSIS .........................................................................................................  27 Time-Ordered Mediation ......................................................................................................  27 Model Construction ..............................................................................................................   27 Variables ..........................................................................................................................  27 Control Variables .............................................................................................................  28 Final Models ....................................................................................................................  31  viii Power Analysis and Effect Sizes .....................................................................................  33   CHAPTER 4: RESULTS ...........................................................................................................  34 Model 1: Simple Mediation Model ......................................................................................  34 Model 2: Controlling for ChildÕs Age ..................................................................................  37 Model 3: Full Mediation Model, Controlling for ChildÕs Age and Quality of Life (T2) .....  38 Summary of Findings ...........................................................................................................  39   CHAPTER 5: DISCUSSION .....................................................................................................  42 Strengths and Limitations .....................................................................................................  44 Limitations .......................................................................................................................  44 Strengths ..........................................................................................................................  46 Research, Policy, and Practice Implications ........................................................................  47 Need for Continued Research on the Use of Children .....................................................  47 Policy and Practice ...........................................................................................................  48 Conclusion ............................................................................................................................  50   APPENDICES ............................................................................................................................  51 APPENDIX A: Use of Children Scale .................................................................................  52 APPENDIX B: Child Behavior ............................................................................................  53 APPENDIX C: Quality of Life Scale ...................................................................................  56   REFERENCES ...........................................................................................................................  57         ix LIST OF TABLES  Table 1 Sample Demographics and Characteristics ...................................................................  22  Table 2 Demographics and Characteristics of Target Children .................................................  23  Table 3 Correlations Among Items in the Use of Children Scale (Time 1 Interview) ...............  26  Table 4 Correlations Among Scales for Use of Children (T1), Child Behavior (T1 and T2), Quality of Life (T1 and T3), Psychological Abuse (T1), Physical Abuse (T1), and Relationship to the Abuser (T1) .................................................................................................  30  Table 5 Means and Standard Deviations for Model Scales across Time Points ........................  35  Table 6 Simple Mediation Model (n = 105) ...............................................................................  37  Table 7 Mediation Model Controlling for ChildÕs Age at T1 (n = 105) ....................................  38  Table 8 Mediation Model Controlling for Quality of Life at T1 and ChildÕs Age (n = 105) ....  39     x LIST OF FIGURES  Figure 1 Mediation Model ..........................................................................................................  17  Figure 2 Mediation Model While Controlling for the Mediating and Dependent Variables at T1 ................................................................................................................................................  28  Figure 3 Models 1 and 2 .............................................................................................................  32  Figure 4 Model 3 ........................................................................................................................  32  Figure 5 Model 1 with Coefficients ...........................................................................................  36  Figure 6 Model 3 with Coefficients ...........................................................................................  39    1 CHAPTER 1: INTRODUCTION Intimate partner violence (IPV) has long been determined to be a public health crisis, affecting millions of women and children each year (Black et al, 2011) with costs of IPV exceeding $5.8 billion annually (CDC, 2003).  In a national survey funded by the National Center for Injury Prevention and Control within the Centers for Disease Control and Prevention, it was found that more than one in three women (36%) have been raped, physically assaulted, and/or stalked in their lifetimes by an intimate partner, including current and former husbands, partners, boyfriends or dates (Black et al., 2011). The same survey found that intimate partners have physically assaulted one out of every four women in the United States. Of the one in five women who have been raped in their lifetimes, over one-half (51%) were raped by an intimate partner. Further, 46% of women have experienced severe psychological abuse over their lifetimes (Breiding et al., 2011).  For the one in six (16%) women who have experienced stalking, two-thirds (66%) were stalked by a current or former intimate partner (Black et al., 2011).   These multiple forms of IPV (e.g., physical, sexual, psychological, stalking, economic) result in significant deleterious outcomes for survivors, including temporary and permanent physical injury (Black et al., 2011; Bonomi et al., 2006; Breiding, Black, & Ryan, 2008; Browne & Williams, 1993; Campbell, 2002; Kwako et al., 2011; Mechanic, Weaver, & Resick, 2008b; Saltzman, Fanslow, McMahon, & Shelley, 2002; Sutherland, Bybee, & Sullivan, 2002; Tjaden & Thoennes, 2000; Tolman, 1989; Wuest et al., 2009; Wuest et al., 2008), economic hardship (Adams, Sullivan, Bybee, & Greeson, 2008; Aguilar & Nightingale, 1994; Anderson et al., 2003; Brewster, 2003; Coker, Smith, Bethea, King, & McKeown, 2000; Davies, Lyon, & Monti-Catania, 1998; Davies & Lyon, 2013; Follingstad, Rutledge, Berg, Hause, & Polek, 1990; Moe  2 & Bell, 2004; Ptacek, 1997; Riger, Ahrens, Blickenstaff, O'Leary, & Maiuro, 2001; Tolman, 1989), and most commonly, mental health issues (Aguilar & Nightingale, 1994; Breiding et al., 2008; Browne & Williams, 1993; Goodman, Koss, Fitzgerald, Russo, & Keita, 1993; Orava, McLeod, & Sharpe, 1996; Sackett & Saunders, 1999; Vitanza, Vogel, & Marshall, 1995). The Characteristics of Intimate Partner Violence Coercive Control  A concept key to defining IPV is the inclusion of coercive control, in which the abusive partner, also defined in the literature as the perpetrator, offender, abuser or batterer, creates a pattern of control through intimidation, isolation, threats, surveillance, and the use of children (Pence & Paymar, 1993; Stark, 2007). This pattern of control is typically punctuated by one or more acts of frightening physical violence, credible threats of physical harm, or sexual assault (Bancroft, 2003; Bancroft & Silverman, 2002; Pence & Paymar, 1993; Saltzman et al., 2002).  Abusers use a multitude of tactics to physically intimidate and harm survivors, including scratching, pushing, and shoving to more severe acts such as strangulation, punching, and stabbing (Black et al., 2011). In addition, sexual violence, abusive sexual contact, and rape are used as tools in attempt to control survivors ((Nemeth, Bonomi, Lee, & Ludwin, 2012; Saltzman et al., 2002)). Psychological and emotional maltreatment further define the abuserÕs employment of power to try to control the victim through humiliation, isolation from friends and family, limiting the victimÕs actions and access to resources, as well as inhibiting a motherÕs ability to parent (Bancroft, Silverman, & Ritchie, 2011; Saltzman et al., 2002; Sutherland et al., 2002). More recently, research on psychological abuse has evolved and researchers have begun to examine more closely specific types of psychological abuse, including economic abuse (Adams, Sullivan, Bybee, & Greeson, 2008; Aguilar & Nightingale, 1994; Anderson et al., 2003;  3 Brewster, 2003; Coker, Smith, Bethea, King, & McKeown, 2000; Davies & Lyon, 2013; Davies, Lyon, & Monti-Catania, 1998; Follingstad, Rutledge, Berg, Hause, & Polek, 1990; Moe & Bell, 2004; Ptacek, 1997; Riger, Ahrens, Blickenstaff, O'Leary, & Maiuro, 2001; Tolman, 1989) and the use of children (Beeble, Bybee, & Sullivan, 2007; Bemiller, 2008; Hardesty & Ganong, 2006; Hardesty, 2002; Harrison, 2008; Kurz, 1996; Mbilinyi, Edleson, Hagemeister, & Beeman, 2007; Moe, 2009; Slote et al., 2005). Increased Lethality When Attempting to Leave the Abusive Partner A common response by those who do not understand the dynamics of IPV is often, ÒWhy doesnÕt she just leave?Ó Research conducted in this area has uncovered just how difficult it can be for survivors to leave an abusive relationship. Reasons include environmental barriers such as lack of money, lack of places to go, homelessness, lack of support from police, from the courts, and from healthcare providers; emotional barriers including the belief that their abusive partner will change; fear that the violence will escalate against their children or other loved ones; and most importantly, escalating threats and abuse (Anderson et al., 2003; Fleury, Sullivan, & Bybee, 2000; Shalansky, Ericksen, & Henderson, 1999). According to Mahoney (1991), separation assault is Òthe attack on the womanÕs body and volition in which her partner seeks to prevent her from leaving, retaliate for the separation, or force her to returnÉIt is an attempt to gain, retain, or regain power in a relationship, or to punish the woman for ending the relationshipÓ (pp. 65-66). For many survivors who leave their abusive partners, the violence simply does not end; in some cases it may escalate against them (Fleury et al., 2000; McFarlane et al., 2005; Tjaden & Thoennes, 1998) as well as their children (Hardesty, 2002; Kurz, 1996; Shalansky et al., 1999). In a study conducted with 135 survivors recruited from a domestic violence shelter, Fleury, Sullivan, and Bybee (2000) found that one in three of  4 the survivors were assaulted post-separation by an ex-partner at least once prior to either reuniting with the partner or before the end of the 2-year study. Survivors attempting to leave or who remained with their violent partners faced threats or experienced increased violence, and increased threats against their children, family, and friends. Hardesty (2002) conducted an extensive literature review regarding separation assault in the context of post-divorce parenting. In her review, she found similarities across studies that having children together made it both difficult for survivors to leave their abusive partners and created a legally-binding context in which batterers had continued physical and emotional access to the survivors. ÒOftentimes, children remain the last link abusers have to their victims through arrangements that guarantee continued access via post-divorce parenting obligationsÓ (Hardesty, 2002, p. 609). In a review of 100,000 womenÕs files from domestic violence shelters, Liss and Stahly (1993) found that 25% reported verbal harassment, and 10% reported physical abuse from their batterers during visitation.  Use of Children As a Form of IPV  An often overlooked form of IPV is the abuserÕs physical and psychological abuse and threats of abuse toward the victimÕs children to accomplish the same end result of power and control (Beeble et al., 2007; Bemiller, 2008; Hardesty & Ganong, 2006; Hardesty, 2002; Harrison, 2008; Kurz, 1996; Mbilinyi et al., 2007; Moe, 2009; Slote et al., 2005). There are a number of forms through which abusers use the children to control their mothers. One strategy is to involve womenÕs children in the abuse, creating a web of deceit that potentially pits family members against each other, which can break down the bond between women and children. These tactics include undermining the womanÕs authority by stopping her from disciplining the children, preventing her from caring for her children, using the children as tools or pawns to  5 harass women by harming or threatening to harm them, or ruining the childrenÕs personal items to punish their mother. These tactics can happen while the woman is still in a relationship with her batterer and can often continue after she attempts to end the relationship, sometimes escalating as a result. Many batterers approach their parenting through a traditionally patriarchal lens both during and post-separation from their victims, viewing their children as extensions of themselves or as property with the need to maintain control over them (and their partners), rather than acting as guides and nurturers (Arendell, 1992; Bancroft & Silverman, 2002; Bancroft et al., 2011; Edleson & Williams, 2007; Hayes, 2012, 2015; Holt, 2015; Jaffe, Johnston, Crooks, & Bala, 2008; Maddox, 2015; McMahon & Pence, 1995). In order to preserve control and maintain the secret of the abuse within the family, batterers may also isolate the children within the home by limiting their access to contacts outside the home (Rossman, Hughes, & Rosenberg, 2000).  Batterers also use the children as a way to stay in womenÕs lives, keep track of them, and intimidate them by manipulating the children to disclose Òguarded informationÓ about their motherÕs residence, employment, or location of family members or friends (Bancroft, 2003; Bancroft & Silverman, 2002; Beeble et al., 2007; Mbilinyi et al., 2007; Tubbs, Williams, & Edelson, 2007) (Maddox, 2015). Mbilinyi and colleagues (2007) found that 79% of their sample of 111 women reported that the batterer used at least one of her children as a tool or pawn against her in order to continue the harassment. In McCloskeyÕs (2001) study of 363 women, 65% who were survivors of IPV also experienced batterersÕ threats to harm or take the children away, an occurrence that was significantly higher than for women who did not experience IPV.  In one of the few studies which has addressed how batterers use children against their mothers, Beeble and colleagues (2007) found that in a sample of 156 women, 88% experienced the batterer using their children in at least one way to coerce or harm them, including 70%  6 stating the batterer used their children to stay in their lives and continue the abuse. Further, 44% of the women said that the batterers used the children to frighten them. Additionally, they examined the connection between the child-abuser relationship and the use of children and found that biological fathers engaged in more use of the children against their mothers than did stepfathers, father-figures, and non-father figures. Similar to the change in abuse tactics some survivors experience when attempting to leave or after leaving batterers, the way abusers use the womenÕs children post-separation may change as well. Tactics include threatening or using physical and sexual violence against the children; keeping the children longer than custody allows or refusing to return the children after visits; abducting or threatening to abduct the children; using children to justify breaking no-contact orders; withholding information about the childrenÕs social, emotional, or physical needs; intentionally setting contradicting rules for the children; discrediting and degrading the survivor as a mother by falsely accusing her of bad parenting, cheating, using drugs, or being Òcrazy;Ó having irregular visitation schedules and demanding visitation with the children within their own schedules; withholding child support, insurance, medical, or basic-expense payments; neglecting or endangering the children when they are with him, or using violence in front of them; ignoring the childrenÕs schedules or their identities; coercing the children to form an alliance with the batterer against their mother; or isolating the children from their mother by not allowing them to speak or contact her during their visitation with the abuser (DeKeseredy & Schwartz, 2009; Greif & Hegar, 1993; Hayes, 2012, 2015; Jaffe et al., 2008; Johnston & Girdner, 2001; Programs, 2013). In a study of 339 women sampled from the Chicago WomenÕs Health Risk Study, Hayes (2015) found that overall, survivors who are separated from their abusers are in fact at an  7 increased risk of threats of Òindirect abuseÓ involving their children. Specifically, when examined by separation status (still together versus not together), survivors who were separated from their abusive partners experienced more threats of harm to their children. There was no significant difference between the two groups on the likelihood of having experienced the abuserÕs threats to take the children, although nearly one quarter of the respondents reported these threats (separated = 23.47%; non-separated = 23.71%). The Impact of Abuse on Survivors Physical Outcomes  Numerous research studies have found that IPV leads to a wide range of negative physical outcomes, resulting from women being slapped, punched, kicked, or thrown to being scalded, cut, strangled, smothered, stabbed or shot (Kwako et al., 2011). More than 1 in 7 women have reported having experienced an injury due to physical violence (Black et al., 2011). Injuries range from bruises, cuts, black eyes, memory loss, concussions, broken bones, and miscarriages to more permanent injuries including scarring, partial loss of hearing or vision, traumatic brain injuries, sexually transmitted diseases, stab wounds, or even death (Bonomi et al., 2006; Browne & Williams, 1993; Mechanic et al., 2008b; Sutherland et al., 2002; Tjaden & Thoennes, 2000; Tolman, 1989; Wuest et al., 2009; Wuest et al., 2008). Long-term outcomes include more visits to health providers over the lifetime, more hospital stays, a longer duration of hospital stays, as well as long term adverse health consequences, including neurological issues such as fainting and seizures, higher than average symptoms associated chronic diseases including gastrointestinal disorders, diabetes, cardiac symptoms such as heart attacks, heart disease, and high blood pressure, and gynecological problems (Black et al., 2011; Breiding et al., 2008; Campbell, 2002). Women who are pregnant  8 and involved with an abusive partner are at a heightened risk of severe physical violence, which can result in especially serious outcomes, including sexually-transmitted diseases, urinary-tract infections, substance abuse, depression, and preterm delivery (Campbell, 2002). The risk of IPV occurrence is heightened at more advanced stages of pregnancy, when women are less able to protect themselves (Browne & Williams, 1993; Saltzman et al., 2002). Psychological Outcomes Short- and long-term psychological outcomes for women who have been repeatedly assaulted by their partners mirror typical reactions of survivors of other traumatic events. Both the timing of the abuse and the duration of the abuse have an impact on the severity of these outcomes (Bonomi et al., 2006). Survivors of repeated assaults often have high levels of depression, suicide ideation, and suicide attempts, as well as chronic fatigue and tension, intense startle reactions, disturbed sleeping and eating patterns, and nightmares (Aguilar & Nightingale, 1994; Breiding et al., 2008; Browne & Williams, 1993; Goodman et al., 1993; Orava et al., 1996; Sackett & Saunders, 1999; Vitanza et al., 1995).   One of the most common negative mental health outcomes experienced by survivors is depression (Bonomi et al., 2006; Dutton, Green, & Kaltman, 2006; Mechanic, Weaver, & Resick, 2008a; Nathanson, Shorey, Tirone, & Rhatigan, 2012; Pico-Alfonso, 2005; Theran, Sullivan, Bogat, & Stewart, 2006). In the most recent systematic review and meta-analysis examining the relationship between IPV and depression, Beydoun and colleagues (Beydoun, Beydoun, Kaufman, Lo, & Zonderman, 2012) reviewed published research articles from 1980 Ð 2010, finding that women who reported IPV were 3.26 times higher than those who did not report IPV to have a major depressive disorder. In an earlier meta-analysis of survivors of IPV, Golding (1999) discovered similarly high prevalence rates of depression (48%) among survivors.  9 The types of abuse women experience typically are co-occurring (Messing, Thaller, & Bagwell, 2014) and increase in amount and severity over time (Browne & Williams, 1993; Sutherland et al., 2002). As a result, it can be difficult to parse out which forms of abuse impact which physical and psychological sequelae. The stress of repeated abuse endangers both a womanÕs psychological and physical health. This stress, compounded with the ongoing physical and psychological abuse, can result in physiological responses to stress that can ultimately be linked with long-term health problems. The mediating effects of this stress and the resulting depression add to womenÕs already long list of physical and mental health problems resulting from the physical abuse (Becker, Stuewig, & McCloskey, 2010; Black et al., 2011; Bonomi et al., 2006; Browne & Williams, 1993; Martin et al., 2008; Sutherland et al., 2002). In a large-scale study based on a survey of 3429 women assessing for IPV exposure and health outcomes, Bonomi and colleagues (2006) found that survivors of IPV were more likely to report depressive symptoms with an increased likelihood of reporting severe depressive symptoms. The survey-based study also proposed that the timing and extent of abuse impacted the level of depressive symptoms, indicating that survivors presented higher levels of depressive symptoms closer to the times of abuse, and survivors who experienced abuse over longer periods of time were more likely to report higher levels of depressive symptoms. Theran and colleagues (2006) also assessed for the impact of IPV on survivorsÕ well-being, which was operationalized to include depression and stress.  Their sample included a total of 398 women, with 52% of the women having been abused by an intimate partner or ex-partner in the prior 6 months. Similar to Bonomi et al. (2006), they found that survivors of IPV reported more depressive symptomology than those who had not been abused. They also found that both  10 physical and psychological abuse contributed significantly and independently to the variance in survivorsÕ depressive symptomology. Mechanic, Weaver, and Resick (2008) also found the link between chronicity and severity of IPV victimization and the high rates of severe mental health symptoms for survivors, including depression and PTSD. The 413 participants in this study were survivors who were seeking help from community agencies serving battered women, who had experienced a minimum of two severe or four minor incidents of violence in the preceding year. In the study, they examined the impact of physical abuse versus psychological abuse and stalking and found that psychological abuse uniquely predicted depression symptoms for the survivors over and above physical abuse. Nathanson and colleagues (2012) had comparable findings, indicating that increased psychological abuse was associated with increased depression while physical abuse was unrelated to symptoms of depression. A study most recently conducted by Estefan, Coulter, and VandeWeerd (2016), found from a sample of survivors recruited from a community-based IPV intervention program, that there is a longitudinal effect between survivorsÕ reports of frequent emotional abuse in the past and self-reported long-term depression. Recently, researchers have taken further notice of the threat of harm and use of children as forms of psychological abuse and have examined the possible link to survivor-mothersÕ mental health (Ahlfs-Dunn & Huth-Bocks, 2016; Rivera, Sullivan, Zeoli, & Bybee, 2016). In a longitudinal study of 40 women who were engaged in the legal process of child custody with their abusive ex-partners, it was found that in addition to physical and psychological abuse contributing to the mental health symptoms of survivors, the threat of harm and/or harm to their children by the abuser additionally contributed to womenÕs depression and PTSD symptoms (Rivera et al., 2016). Ahlfs-Dunn and Huth-Bocks (2016) found similar connections between the  11 use of children as a form of abuse and survivorsÕ overall mental health, including depression, for survivors who were parenting very young children. While both of these studies suggest a link between the use of children and a variety of mental health-related outcomes, the samples are bounded by specific experiences including being in the process of leaving an abusive partner (Rivera et al., 2016) and parenting young children (Ahlfs-Dunn & Huth-Bocks, 2016). Additionally, both studies defined the use of children differently. Rivera and colleagues (2016) used a broad scale that included harm to, threats to harm, and use of children without disentangling the specific harm to the children. Alhfs-Dunn and Huth-Bocks (2016) operationalized the use of children as the ways in which the abuser intentionally interfered with the survivorsÕ parenting. Quality of Life Being victimized by a partner or ex-partner can also decrease a survivorsÕ overall quality of life (QOL; (Adeodato, Carvalho, Siqueira, & Souza, 2005; Beeble, Bybee, Sullivan, & Adams, 2009; Bybee & Sullivan, 2002; Laffaye, Kennedy, & Stein, 2003; Leung et al., 2005; Sullivan, 2003; Sullivan & Bybee, 1999). Although the link between IPV and survivorsÕ QOL has not been extensively studied, those studies that have been conducted have operationalized quality of life variably and in broad terms, linking it to specific physical and psychological health outcomes including stress and PTSD, social support, and an overall sense of well-being (Laffaye, Kennedy, & Stein, 2003). Literature to date that has examined the relationship between QOL and IPV has used different measures for quality of life. Some of the measures that have been used include the adapted scale from Andrews and Withey (1976; Beeble et al., 2009; Bybee & Sullivan, 2002; Sullivan & Bybee, 1999; Tan, Basta, Sullivan, & Davidson, 1995), and health-related scales that inform individualsÕ quality of life, including The Medical Outcome Study  12 Short Form (SF-36) scale (Laffaye et al., 2003; Ware, Snow, Kosinski, & Gandek, 1993), and the World Health Organization Quality of Life (WHOQOL-Bref instrument (Harper, 1998; MacMillan et al., 2009). In a communityÐsampled study of 70 women (40 abused and 30 non-abused), Laffaye and colleagues (2003) used the Short-Form health Survey (Ware, Snow, Kosinski, & Gandek, 1993) to assess participantsÕ quality of life in eight different areas including physical functioning, role limitations due to physical health, bodily pain, general health, vitality, social functioning, role limitations due to emotional problems, and mental health. Participants were assessed at one time point and comparisons were made between the group of women who had been abused and those who had not, finding that survivors of IPV reported significantly lower levels of health-related quality of life than those participants who had not been abused.  In a study examining the 2-year trajectories of 160 help-seeking survivors of IPV, including their experiences of abuse, researchers examined whether QOL (using an adapted version of the Andrews and Withey scale; 1976) was mediated or moderated by social support (Beeble, Bybee, Sullivan, & Adams, 2009). Among the survivors of IPV, those who had experienced higher levels of psychological abuse had lower QOL scores. As psychological abuse decreased over time, survivorsÕ QOL increased, while changes in physical abuse did not have the same impact. When social support was entered in to the models as a mediator, Beeble and colleagues identified that social support mediated the effects of change in psychological abuse on QOL and depression for within-person change. Finally, Wittenberg and colleagues (2007) conducted a focus group-based study with 40 survivors of IPV designed to analyze a number of health-related quality of life measures. From the eight focus groups, they learned that many survivors talked about their quality of life in  13 relation to their childrenÕs quality of life, suggesting that health-related QOL measures should be more ÒholisticÓ by including questions related to childrenÕs and the family unitÕs quality of life as well.  The Impact of IPV on Children It has been estimated that 16.3% of all children 17 years of age or younger have witnessed physical violence within their lifetimes (Finkelhor, Turner, Ormrod, & Hamby, 2009). This translates to approximately 15 million children, 7 million of whom live in a household where severe violence occurs (McDonald, Jouriles, Ramisetty-Mikler, Caetano, & Green, 2006).  In a qualitative study of children staying in domestic violence shelters, almost three-quarters of the children (71%) witnessed severe physical violence, reporting that they saw pushing or shoving, kicking, biting, punching, objects being thrown at, or a weapon being used against their mothers (DeBoard-Lucas & Grych, 2011).  Children witness IPV in a multitude of ways, both directly and indirectly. This can include seeing the abusive acts, hearing the abuse, being used as a pawn by the abuser, witnessing the aftermath of a violent event, and/or most tragically witnessing a parentÕs homicide (Beeman & Edleson, 2000; Carter & Schechter, 1997; Mbilinyi et al., 2007). According to Sullivan and colleagues (Sullivan, Juras, Bybee, Nguyen, & Allen, 2000), children whose mothersÕ batterers were their biological fathers witnessed more violence against their mothers than children whose mothersÕ batterers were non-father figures or stepfathers. Observing abuse happens both directly and indirectly and can include seeing or hearing the abuse, and seeing the outcomes of the abuse such as witnessing bruises on their mother, broken furniture in their home, or becoming aware of the violence as a result of a family member telling them about it (Graham-Bermann & Edleson, 2001).  14 In addition to exposure to the IPV, there has long been documented co-occurrence of IPV and child maltreatment, including abuse, neglect, and child sexual abuse (Bancroft, 2003; Dick, 2006; Holden, Barker, Appel, & Hazlewood, 2010; Kitzmann, Gaylord, Holt, & Kenny, 2003; McCloskey, 2001; McGee, Wolfe, & Wilson, 1997). Children who live in homes were there is violence are at an increased risk for both physical and emotional abuse (Edleson, 1999b; Hamby, Finkelhor, Turner, & Ormrod, 2010; Rizo, Macy, Ermentrout, & Johns, 2011).  In Holt, Buckley, and WhelanÕs (2008) literature review, the co-occurrence between IPV and child abuse was found to occur in 45-70% of the families across studies. A large-scale study conducted by Hamby and colleagues (2010) examined the link between witnessing IPV and maltreatment including child sexual abuse using data from the National Survey of ChildrenÕs Exposure to Violence. Of the 4549 youth who participated in the study, they found that youth who witnessed IPV (34%) were three to nine times more likely to be maltreated than those who had not witnessed IPV.   Children who are raised in abusive home environments are at risk in a myriad of ways, typically relating to their social-emotional development (Gewirtz & Edleson, 2007). Infants and toddlers exposed to IPV are at risk for higher levels of irritability, sleep disturbances, emotional distress, and insecurities around being left alone (Edleson, 1999a; Holt, Buckley, & Whelan, 2008; Lundy & Grossman, 2005; Osofsky, 2003). In a study examining the risk and protective factors of toddlers and young children aged 2 - 4, Martinez-Torteya and colleagues (2009) found that young children exposed to IPV were 3.7 more likely to develop internalizing and externalizing behaviors than those who had not been exposed. Holt and colleagues (2008) also found that pre-school age children who have witnessed IPV are at a greater risk for Òmore  15 behavioral problems, social problems, post-traumatic stress symptoms, greater difficulty developing empathy, and poorer self-esteem than non-witnesses,Ó (p. 802).   As children grow older, living in an abusive home and witnessing IPV puts them at risk for poor social skills, aggressiveness, acting out, drug use, peer difficulties, depression, PTSD symptoms, suicide attempts, and decreased learning potential (Holt et al., 2008; Lundy & Grossman, 2005; Moore & Pepler, 1998). In a study of 112 children sampled from the community aged 6-13, Bauer (2006)  found that children exposed to IPV were 3.1 times more likely to exhibit externalizing behavior problems than those who were not exposed to IPV. Living in a home where one parent is abusing the other can also have deleterious effects for adolescents, including a possible increase of victimization in their own relationships (Holt et al., 2008; Levendosky, Huth-Bocks, & Semel, 2002), PTSD and depression symptoms, suicide attempts, and delinquency, including poly-drug use (Brockie, Dana-Sacco, Wallen, Wilcox, & Campbell, 2015; Moylan et al., 2010). In a longitudinal study based on data collected by the National Survey of Child and Adolescent Well-Being, Holmes (2012) followed 730 children over the course of eight years with the first time point of data being collected when the children were between birth and three years old. Holmes found significant connections between exposure to IPV and aggressive behavior problems at multiple time points. Both age of exposure and frequency of exposure to severe forms of IPV related to an increase in aggressive behavior over time. Current Study There are few studies examining the use of children as an abusive tactic over and above traditionally recognized forms of abuse. One study to date has explored the longitudinal impact of the harm of and use of children on survivorsÕ depression (Rivera, Sullivan, Zeoli, & Bybee,  16 2016), but this small study only examined the experiences of survivors post-separation involved in the family court system regarding custody. Another study that examined the impact of abusersÕ use of children on survivor outcomes operationalized Ôuse of childrenÕ only in terms of the abuserÕs interference with the survivorÕs ability to parent young children under the age of three (Ahlfs-Dunn & Huth-Bocks, 2016). The proposed study is designed to expand this scant literature by examining whether the batterersÕ use of the children impacts survivorsÕ quality of life over time. The study will also be the first to examine whether this form of abuse impacts childrenÕs behaviors. Finally, the study plans to explore survivorsÕ perceptions of their childrenÕs behaviors as a mediator in the relationship between the use of children and survivorsÕ quality of life.   The impact of abuse on survivorsÕ well-being is known to change over time (Beeble, Bybee, Sullivan, & Adams, 2009; Browne & Williams, 1993; Sutherland et al., 2002; Estefan, Coulter, & VandeWeerd, 2016; Rivera et al., 2016).  Building upon this knowledge, the proposed study will examine whether abuse of the children at one time point impacts survivorsÕ well-being and childrenÕs behaviors at later time points.  The proposed study involves secondary data analysis of a longitudinal research study that included multiple time points over two years.  Study Hypothesis With the understanding that it may take time to see the effects of abuse, this study will look at three time points. It is hypothesized that women who report higher levels of the abuserÕs use of the children as an abuse tactic at Time 1 will report higher behavioral problems in their children four months later (Time 2). In turn, higher behavioral problems of the children will lead  17 to lower quality of life for survivors four months after that (Time 3). See Figure 1 for a depiction of the basic model representing the hypothesis. Figure 1  Mediation Model.     X M Y T1 T2 T3 cÕ a b X = Use of Children at  Time 1 M = ChildrenÕs behavior at Time 2 Y = QOL at Time 3  18 CHAPTER 2: METHOD This study uses a subset of data from a longitudinal study that followed families over the course of 24 months had experienced IPV. Families were randomized into one of two groups; those in the intervention group were given the opportunity to participate in a 4-month intervention providing mothers with trained paraprofessional advocates and providing children with a weekly support and education group. Women and their children were interviewed at six time points -- every four months during year 1 and every six months during year 2. Due to the variability in the gaps between the interviews for years 1 and 2, three interviews -- all four months apart -- were used to test the study hypothesis. Because survivors were recruited into the study at a point of crisis and high incidence of abuse (see eligibility criteria below), their baseline interviews were excluded from the model and interviews were used from the 4-month, 8-month, and 12-month time points. Procedures Recruitment Women were recruited into the study from three locations within the community: a community-based agency providing short-term support to victims of domestic violence following a police intervention (43.8%); a domestic violence shelter (25.7%); or the county prosecutorÕs personal protection order office (30.5%). Eligibility Eligibility criteria for the study included the following: 1) women had to have experienced at least one form of physical abuse in the four months prior to their first interviews; and 2) women had to have at least one child between the ages of 5 and 12 who was also interested in participating in the research study.    19 Interviews Women who agreed to participate in the study were given the choice of where to be interviewed based on their level of comfort and need for safety. Women recruited from the residential shelter were not interviewed until after they had left shelter in order to avoid interviewing women during a time of crisis. Women were paid $20 for participating in the first interview and  $60, $70, $80, $90 and $100 for their subsequent interviews. All interviewers were highly-trained, having gone through a standardized and formal interviewer training specifically for interviewing survivors of intimate partner violence. Participants  Of the original sample of 160 women, women in this study were included if they completed interviews at the 4-month, 8-month, and 12-month time points. One woman was excluded because the age of the target child (3 years old) fell far outside the designated age range of five to thirteen years old. Four women were excluded based on the assailantÕs lack of relationship to the target child (he was not a father, a step-father, or a father figure) combined with his having no contact with the child at the first time point. Finally, any women who did not complete all of the relevant scales were removed from the data set. The final sample for this study, then, was 105 IPV survivors. Demographics  Survivors. Of the 105 women included in the study, their ages ranged from 22 to 49 years old (M = 32.31; SD = 6.17). Almost half of the women identified as white or Caucasian (n = 54, 51.4%); one-quarter identified as Black or African American (n = 30; 28.6%); 9 (8.6%) identified as Hispanic or Latina; 8 women (7.6%) identified as multiracial; 2 women (1.9%) identified as Native American; and one woman (1.0%) identified as Sudanese. One woman  20 (1.0%) did not give her race or ethnicity. WomenÕs educational levels varied ranging from less than high school to a bachelorÕs degree. Sixty-one (58.1%) of the women had completed at least some education past high school, 27 women (25.7%) completed high school, and 17 women (16.2%) had not completed high school.  The average number of children a woman had was 3.02 (SD  = 1.37; range = 1 - 8 children). The majority of the women had two (n = 35; 33.3%) or three (n = 36; 34.3%) children. Twenty-three (21.9%) of the women had between four and five children. A total of 317 children were represented in this study, 183 (57.7%) of the children were between the ages of 5 to 12 years old.  At the time of their first interview the majority of the women identified that they were planning on ending the relationship with the assailant (n = 87; 82.9%). Among those continuing the relationship with the assailant, 15 of the women (14.3%) indicated they were living together and 3 women (2.9%) indicated they were staying together but not living together. The demographics and characteristics of the women involved in the study are presented in Table 1. Target children. Within each family, one child between the ages 5 and 12 was randomly chosen by the research team to be included in the study.1 Survivors were asked specific questions about the target children, the target children were also interviewed themselves, and they took part weekly support and education groups. Among those target children, the average age was 8.94                                                 1 It is important to note that there were four target children who were four years old and two target children who were thirteen years old at the time of recruitment. The research team decided to include these children in the study as target children as their birth dates were close to the cut-off ages.  21 years old (SD = 2.24), with a range of 4 to 13 years old. Twenty-five (23.8%) of the children were age six or under, over one-half (n = 57; 54.3%) of the children were between the ages of 7 and 10 years old, and 23 of the children (21.9%) were 11 to 13 years old. Nearly one-third of the children were identified as white (n = 39, 37.1%) and one-third were identified as Black or African American (n = 32, 30.5%) by their mothers, 25 children (23.8%) were identified as multiracial, six children (5.7%) were Hispanic or Latin@, one child (1.0%) was Sudanese, and two children (1.9%) did not have their race or ethnicity identified.  Of the 125 target children, all had had contact with the assailant at least once in the four months prior to the first interview. Their relationships to the assailant, as identified by their mothers, included: (1) biological father for 45 (42.9%) of the children, (2) stepfather for 21 (20.0%) of the children, (3) father figure for 26 (24.8%) of the children, and (4) not a father figure for 13 (12.4%) of the children. Stepfathers were those assailants having been legally married to the childrenÕs mother. Father-figures were identified as those who played a significant role in the childrenÕs lives. Non-father figures were those who did not play a significant role in the childrenÕs lives. Table 2 presents the demographics and characteristics of the target children involved in the study.    22 Table 1  Sample Demographics and Characteristics  Demographics Count Percentage Race   Black/African American 30 28.6 White 54 51.4 Hispanic or Latin@ 9 8.6 Native American/First Nation 2 1.9 Sudanese 1 1.0 Multi Racial 8 7.6 Missing 1 1.0 Total Number of children   1 child 7 6.7% 2 children 35 33.3 3 children 36 34.3 4 children 12 11.4 5 children 11 10.5 6 children 1 1.0 7 children 1 1.0 8 children 2 1.9 Educational level at Pre   Less than High School 17 16.2% High School Graduate 27 25.7 Trade School Graduate 7 6.7 Some College 39 37.1 Associates Degree 12 11.4 Bachelors Degree 3 2.9 Recruited from at Pre    Shelter 27 25.7 Police Intervention 46 43.8 PPO/Court 32 30.5     23 Table 2 Demographics and Characteristics of Target Children Demographics Count Percentage Target Child Race   Black/African American 33 30.6% White 40 37.0 Hispanic or Latin@ 6 5.6 Sudanese 1 0.9 Multiracial 26 24.1 Missing 2 1.9 Target Child Age   4 years old 4 3.7% 5 years old 8 7.4 6 years old 13 12.0 7 years old 15 13.9 8 years old 17 15.7 9 years old 15 13.9 10 years old 13 12.0 AssailantÕs Relationship to Target Child at Baseline    Biological father 45 41.8% Stepfather 22 20.4 Father Figure 28 25.9 Not a father figure 13 12.0  Measures AssailantÕs Relationship to the Survivor and Target Child Women were asked at the first interview if they were continuing or ending their relationship with the assailant. They were also asked to indicate if the abuser was the childÕs biological father, step-father, a father figure, or a non-father figure. Intimate Partner Violence  Physical abuse. A modified version of the Conflict Tactics Scale (CTS; Straus, 1979; Sullivan & Bybee, 1999) was used to ask women about the abuse perpetrated by the assailant. Women were asked about multiple types of physical abuse they may have experienced over the last four months (e.g., ÒHas the assailant pushed or shoved in you in the last four months?Ó).  A  24 seven-point Likert scale was used (1 = never to 7 = more than 4 times a week). CronbachÕs alpha for the original study was 0.89 (M = 1.36; SD = 0.95). Emotional abuse. A shortened version of the Index of Psychological Abuse (IPA; (Sullivan, Tan, Basta, Rumptz, & Davidson, 1992) was used to measure the psychological/emotional abuse women experienced in the prior 4-month period. The 24-item index assessed women on a wide range of items related to emotional and psychological abuse (e.g., ÒIn the last four months has the assailant called you names?Ó). Women were asked to respond using a four-point Likert scale (1 = never to 4 = often). CronbachÕs alpha in the original study was 0.89 (M = 2.49; SD = 0.61). Use of children. A seven-item scale was created for this study in order to examine the assailantÕs use of the children in order to control their partner or ex-partner. Items in this scale included whether the assailant had used the children to harass or intimidate the woman, to stay in her life or to keep track of her, or to frighten her (See Appendix A for the full scale). Women were also asked if the assailant had tried to turn the children against her or had tried to convince the children that she should take him back. A four-point Likert scale was used (1 = never to 4 = often). CronbachÕs alpha for this scale was 0.87 (M = 2.27; SD = 0.92). Of the 105 women in the final sample, eighty-four (81.0%) of the women said they had experienced the abuser using their children as an abuse tactic at least once in the previous four months. Table 3 presents the inter-item correlation for the use of children scale at the 4-month, or time one (T1) interview.  ChildrenÕs Behavior A total of 45 Items were used to create a scale for this study, based on a combination of items in the Eyberg Child Inventory (36 items; (Eyberg & Ross, 1978; Robinson, Eyberg, & Ross, 1980) and AchenbachÕs Child Behavior Checklist (9 items; CBCL; (Achenbach &  25 Edelbrock, 1983) to create a scale specific to this study (see Appendix B for a full list of items). Only mothers completed this measure, examining the range of aggressive and withdrawal behaviors of children (e.g., ÒMy child has temper tantrumsÓ;  ÒMy child feels worthless or inferiorÓ). Survivors were asked how often their children behaved in specific ways using a Likert scale (1 = never to 7 = always). CronbachÕs alpha for this scale was 0.94 (M = 2.58; SD = 1.10). Quality of Life Items were adapted from Andrews and Withey (1976) to create a 9-item scale to assess the womenÕs perceived quality of life. Women were asked to respond to each item (e.g, ÒHow do you feel about the responsibilities you have for members of your family?Ó) using a 7-point Likert scale (1 = very unhappy to 7 = very happy). This scale has been used in many prior studies on IPV (Beeble, Sullivan, Bybee, & Adams, 2009; Bybee & Sullivan, 2002; Sullivan & Bybee, 1999; Tan, Basta, Sullivan, & Davidson, 1995). CronbachÕs alpha for this scale was 0.86 (M = 4.29; SD = 1.06). See Appendix C for the entire quality of life scale. Intervention Condition Families were randomized into a control group or an intervention group. The women and children in the intervention group had the opportunity to work with an advocate over the course of 10 weeks to meet their needs and the children had the opportunity to attend a 10-week support and education group. Sixty-two women (59.0%) were in the control group and 43 (41.0%) of the women were in the intervention group.    26 Table 3  Correlations Among Items in the Use of Children Scale (Time 1 Interview) Did the batterer É 1 2 3 4 5 6 1. Use the children to stay in your life       2. Use the children to harass you .56***      3. Use the children to intimidate you .53*** .80***     4. Use the children to keep track of you .38*** .60*** .57***    5. Use the children to frighten you .38*** .61*** .63*** .43***   6. Try to turn the children against you .53*** .54*** .49*** .41*** .47***  7. Try to convince the children you should take him back .57*** .38*** .32*** .36*** .24*** .27** Note. ***p<.001; **p<.01   27 CHAPTER 3: ANALYSIS Time-Ordered Mediation The longitudinal nature of this study allows for examining predictive, rather than simply correlative, relationships. Mediation is conceptualized as being a process that occurs over time (Collins, Graham, & Flaherty, 1998). In order to identify a mediation (Baron & Kenny, 1986),  the following must occur: 1) There must be a significant total effect of the independent variable (X) on the dependent variable (Y). 2) There must be a significant effect of the independent variable (X) on the mediating variable (M). 3) The effect of the mediating variable (M) on the dependent variable (Y) must be significant when controlling for the independent variable (X).  4) The effect of the independent variable (X) on the dependent variable (Y) when controlling for the mediating variable (M), must be smaller than the total effect of X on Y (as found in number 1). Model Construction Variables The use of children scale at the 4-month interview (T1) was used as the predictor variable in the model. ChildrenÕs behavior four months later (T2), as assessed by their mothers, was used as the mediating variable. WomenÕs quality of life (QOL) at the twelve-month interview or time 3 (T3) interview was used as the dependent variable in the model. Both the independent and mediating variables were assessed for their predictability of the dependent variable, both alone and collectively.  28 Control Variables A number of control variables were also examined as part of the model based on prior research showing their relation to survivorsÕ quality of life or to the use of the children. In the time-ordered mediation model, the measures of the mediator (child behavior) and dependent variable (quality of life) at the first time point of the model (T1) were entered into the regression model as control variables (see Figure 2). Additional control variables were also examined as part the model, including the abuserÕs relationship to the children, the childrenÕs age, physical abuse, psychological abuse, recruitment site, and intervention condition. Each of these variables was examined individually in the model to determine their impact. Due to the small sample size of the study (n = 105), any additional control variable added took up a large amount of space within the model. As a result, those control variables that had no influence in the model were removed. Decisions to omit variables were based on two criteria: lack of significance and multicollinearity.  Figure 2  Mediation Model While Controlling for the Mediating and Dependent Variables at T1.   M2 Y3 T1 T2 T3 cÕ a b X1 = Use of Children at  Time 1 M1 = ChildrenÕs behavior at Time 1 M2 = ChildrenÕs behavior at Time 2 Y1 = QOL at Time 1 Y3 = QOL at Time 3 X1 Y1 M1  29 Each control variable was entered into the model; variables that showed no significant impact on the relationship between the independent, mediating, and dependent variables were removed. Two variables were recoded into dummy variables and each was examined for impact in the overall model: 1) abuserÕs relationship to the children (biological father, step father, father figure, not a father figure) and 2) recruitment site (shelter, police intervention, and personal protection order office). Variables that had no significant impact included the abuserÕs relationship to the children, the condition assigned to the family at the beginning of the study, recruitment site and the scale of physical abuse.  The second criterion was when a control variable exhibited multicollinearity with at least one of the other variables in the model. As highlighted in Table 4, child behavior at T1, quality of life at T1, and psychological abuse at T1 were highly correlated with the three main model variables. As expected, child behavior at T1 was highly correlated with child behavior at T2 (! = 0.72, p < .001). Yet, when child behavior at T1 was entered into the models it did not influence the impact of child behavior at T2 on the dependent variable (quality of life at T3); the relationship between child behavior at T2 and quality of life at T3 remained significant. Psychological abuse at T1 was correlated with the use of children at T1 (! = 0.63, p < .001) and child behavior at T2 (! = 0.52, p < .001). Further examination of the two scales showed a number of items in the scale of psychological abuse with medium to larger correlations to the use of children scale. Items included: 1) the assailant having punished or deprived the children of something they needed in the past four months (! = 0.41, p < .001); 2) the assailant having threatened to take the kids away in the past four months (! = 0.65, p < .001); 3) the assailant having threatened to end the relationship in the past four months (! = 0.62, p < .001); and 4) the    30 Table 4  Correlations Among Scales for Use of Children (T1), Child Behavior (T1 and T2), Quality of Life (T1 and T3), Psychological Abuse (T1), Physical Abuse (T1), and Relationship to the Abuser (T1)  1 2 3 4 5 6 7 8 9 10 1. Use of Children at T1           2. Child Behavior at T1 .37***          3. Quality of Life at T1 -.19* -.45***         4. Psychological Abuse at T1 .63*** .52*** -.48***        5. Physical Abuse at T1 ns ns -.33*** .39***       6. Child Behavior at T2  .22** .72*** -.39*** .47*** ns      7. Quality of Life at T3 -.21* -.37*** .75*** -.37*** -.21* -.40***     8. Biological Father .24* ns ns ns ns ns ns    9. Stepfather ns ns ns ns ns ns ns -.43***   10. Father Figure ns ns ns ns ns ns ns -.50*** -.29**  11. Not a Father Figure ns ns ns ns ns ns ns -.33*** ns -.22* Note. ***p < .001; **p < .01; *p < .05   31 assailant having accused the survivor of being a bad mom in the past four months (! = 0.51, p < .001).  After individual examination of each control variableÕs influence in the model, two control variables remained: the age of the children, and survivorÕs quality of life at the first time point (T1).  Final Models Using three points in time, analyses incorporated survivorsÕ perceptions of their childrenÕs behavior as a mediator. Multiple regressions were conducted to determine if the use of children and survivorsÕ assessment of their childrenÕs behavior have an impact on survivorsÕ quality of life as time passes. Each variable was examined for its predictability of the dependent variables. The order of regressions for the model was as follows: ¥ First regression: use of children at the first time point (T1) predicting childrenÕs behavior at the second time point (T2) four months later. ¥ Second regression: use of children (T1) and childrenÕs behavior (T2) predicting survivorsÕ quality of life at the third time point (T3), eight months following the T1 interview. ¥ Third regression: use of children at T1 predicting survivorsÕ quality of life at T3 (eight months later). Three variations of this model were run. Model 1 was the simple model, which included only the independent (use of children T2), mediating (childrenÕs behavior T3), and dependent (survivorsÕ quality of life T4) variables. Model 2 used the same simple model while controlling for the age of the target children involved in the study. Model 3, similar to Model 2, examined the influence of the use of children as an abuse tactic at T1 through survivorsÕ perceptions of their childrenÕs  32 behavior four months later at T2 on the survivorsÕ quality of life four months after that (T3) while controlling for the age of the target children involved in the study as well as the survivorsÕ quality of life at T1. See Figures 3 and 4 for diagrams of Models 1, 2 and 3.    Figure 3 Models 1 and 2.   Figure 4 Model 3.    X M Y T1 T2 T3 cÕ a b X = Use of Children at  Time 1 M = ChildrenÕs behavior at Time 2 Y = QOL at Time 3 M2 Y3 T1 T2 T3 X1 = Use of Children at  Time 1 M2 = ChildrenÕs behavior at Time 2 Y1 = QOL at Time 1 Y3 = QOL at Time 3 X1 Y1 a c' b  33 Power Analysis and Effect Sizes  Bootstrapping was, used to construct the confidence intervals for power analysis and to detect the indirect effects. In order to accomplish bootstrapping, the PROCESS software was added into the SPSS statistical package to produce accurate and precise confidence interval estimates (Hayes, 2013; Preacher & Selig, 2012).  Fritz and MacKinnon (2007) suggest that with a sample size of at least 71, in order to achieve power of 0.80, the a coefficient (use of children) and b coefficient (childrenÕs behavior) both need to need to be 0.39 using bias-corrected bootstrap. For a larger sample size of n = 115, they suggest the coefficients should be 0.25 and 0.40, interchangeably.    34 CHAPTER 4: RESULTS   Of the 160 families who fit the eligibility criteria for this study, 105 were retained over time. Those families who did not remain in the final analysis were removed due to lack of completion of all of the measures included in the models.   Use of the children as part of the abuse strategy was high and decreased over time. At the first time point, 4 months into the study, 80.0% of the survivors reported at least one incident of the use of the children; at the second time point, 66.7% survivors reported at least one incident; and at the third time point, half of the survivors (50.6%) reported at least one incident of the use of the children1. SurvivorsÕ quality of life remained fairly stable across the three time points, showing a slight increase from the first time point (M = 4.90) to the third time pint (M = 5.03). Survivors also reported that their target childrenÕs negative behavior stayed relatively stable as well, with the level decreasing from the first time point (M = 2.93) to the second time point (M = 2.75; ) and increasingly slightly from the second to third time point (M = 2.77). Paired sample t-tests were conducted to examine the change of the variables across time points. Means, standard deviations, and paired t-test results of the three scales are reported in Table 5. Model 1: Simple Mediation Model  A simple mediation analysis using ordinary least squares path analysis revealed that the use of children as an abuse tactic indirectly influenced survivorsÕ quality of life eight months later as mediated by childrenÕs behavior at a time point between the two. Survivors who reported                                                 1 Although baseline data were not included in the model, it important to note that 90.5% of the survivors reported use of the children at this time point.  35 Table 5 Means and Standard Deviations for Model Scales across Time Points  Time point 1 Time point 2 Time point 3   Scale name M SD M SD M SD t-test  T1 to T2 t-test  T2 to T3 Use of Children 1.91 0.85 1.66 0.84 1.55 0.79 3.70** ns Child Behavior 2.93 1.09 2.75 0.88 2.77 1.07 2.38* ns Quality of Life 4.90 1.01 4.94 1.07 5.03 1.11 ns ns Note. *p < .05; **p < .001. Use of Children ranges from 1 (never) to (often). Child Behavior examines items relating to aggressive and withdrawal behavior ranging from 1 (never) to 7 (always). Quality of life ranges from 1 (very unhappy) to 7 (very happy).   36 high levels of use of the children as an abuse tactic reported higher levels of negative child behavior four months later (a = 0.25), and survivors who reported their children having higher levels of negative behavior at four months also reported having lower quality of life at eight months (b = -0.45). A bias-corrected bootstrap confidence interval for the indirect effect (ab = -0.11) based on 10,000 bootstrap samples was entirely below zero (-0.23 to -0.02). There was little evidence that the use of children as an abuse tactic at the first time point influenced survivorÕs quality of life eight months later independent of its effect on child behavior at four months (cÕ = -0.19, p = 0.12) supporting the hypothesis that the use of children, as mediated by child behavior, impacts a survivorÕs quality of life over time. Coefficients and model descriptions are depicted in Figure 5 and Table 6. Figure 5 Model 1 with Coefficients.    X M Y T1 T2 T3 a = 0.25  X = Use of Children at  Time 1 M = ChildrenÕs behavior at Time 2 Y = QOL at Time 3 cÕ = -0.19  b= -0.45   37 Table 6 Simple Mediation Model (n = 105)   Consequent Antecedent M (Child Behavior at T3)  Y (Quality of Life T4) Coeff. SE p  Coeff. SE p          X (Use of Child T2) a 0.25 0.10 .01 cÕ -0.19 0.12 0.12 M (Child Behavior T3)     b -0.45 0.12 <.001 Constant i1 2.28 0.21 <.001 i2 6.84 0.36 <.001            R2 = 0.06  R2 = 0.18   F (1, 103) = 6.18, p = 0.01  F (2, 102) = 10.91, p <.001  Model 2: Controlling for ChildÕs Age  Based on the initial analyses examining the influence of control variables, it was determined that the model should be explored controlling for the age of the children. When controlling for the childÕs age, the use of the children as an abuse tactic still influenced the dependent variable (quality of life at 12 months) through the mediating variable (child behavior at eight months). When controlling for the age of the children, the use of children as an abuse tactic at the first time point (T1) influenced the childrenÕs behavior four months later (T2; a = 0.25, p = .01). The influence of child behavior at the eight month time point (T2) significantly impacted the survivorÕs quality of life four months later at the twelve month time point (T3) when controlling for the childÕs age (b = -0.45, p < .001), suggesting that the more negative behavior the mother perceived her child to exhibit, the lower her quality of life four months later.  The bias-corrected bootstrap confidence interval for the indirect effect (ab = -0.11) based on 10,000 bootstrap samples was entirely below zero (-0.24 to -0.02). Finally, the use of children as an abuse tactic at the first time point did not impact survivorÕs quality of life eight months later independent of its effect on child behavior at four months (cÕ = -0.20, p = 0.10). As indicated in Table 6, the hypothesis remains supported that child behavior serves as a mediating variable in  38 the relationship between the use of children as an abuse tactic and survivorÕs quality of life eight months later.  Table 7 Mediation Model Controlling for ChildÕs Age at T1 (n = 105)   Consequent Antecedent M (Child Behavior at T2)  Y (Quality of Life T3) Coeff. SE p  Coeff. SE p          X (Use of Child T1) a 0.25 0.10 0.01 cÕ -0.20 0.12 0.10 M (Child Behavior T2) - - - - b -0.45 0.11 p <.001 Constant i1 2.24 0.41 <.001 i2  7.51 0.53 p <.001 Control Ð (ChildÕs age)  0.005 0.04 0.91  -0.10 0.04 0.03            R2 = 0.06  R2 = 0.21   F (2, 102) = 3.07, p = .05  F (3, 101) = 9.18, p <.001  Model 3: Full Mediation Model, Controlling for ChildÕs Age and Quality of Life (T2)  The final model was created controlling for both the survivorÕs quality of life at the first time point (T1) as well as the childÕs age. Coefficients and model descriptions are presented in Figure 6 and Table 8. When controlling for the survivorÕs quality of life at T1 as well as the target childÕs age, the use of children trended toward significantly impacting the childÕs behavior four months later (a = 0.17, p = 0.08), suggesting that the more often children are used to harm their mothers, the more negative behavior they may exhibit four months later. When controlling for the survivorÕs quality of life at T1, the childÕs behavior at the eight-month time interval (T2) no longer significantly impacted the survivorÕs quality of life at T3 (b = -0.13, p = 0.14). The bias-corrected bootstrap confidence interval for the indirect effect (ab = -0.02) based on 10,000 bootstrap samples ranged from below to above zero (-0.09 to 0.005), indicating no mediation. Finally, the use of children as an abuse tactic at the first time point did not impact the survivorÕs quality of life eight months later independent of its effect on child behavior at four months (cÕ = -0.09, p = 0.29). Thus, when controlling for the survivorÕs quality of life at T1 as well as the  39 childrenÕs age, the mediation of the use of children as an abuse tactic by the childÕs behavior four months later no longer had a significant impact on survivorsÕ quality of life at the twelve month time point (T3), suggesting support for the null hypothesis. Figure 6 Model 3 with Coefficients.  Table 8 Mediation Model Controlling for Quality of Life at T1 and ChildÕs Age (n = 105)   Consequent Antecedent M (Child Behavior at T2)  Y (Quality of Life T3) Coeff. SE p  Coeff. SE p          X (Use of Child T1) a 0.17 0.10 0.08 cÕ -0.09 0.09 0.29 M (Child Behavior T2) - - - - b -0.13 0.09 0.14 Constant i1 4.39 0.65 <.001 i2 1.97 0.71 0.006 Control Ð (Quality of Life T2)  -0.34 0.08 <.001  0.75 0.08 p <.001 Child Age  -0.03 0.04 0.35  -0.01 0.03 0.76   R2 = 0.19  R2 = 0.58   F (3, 101) = 7.86, p <.001  F (4, 100) = 34.80, p <.001  Summary of Findings  As indicated in the first two models, there is some evidence that childrenÕs behavior serves as a mediator between the relationship between the use of children as a tactic of abuse and M2 Y3 T1 T2 T3 X1 = Use of Children at  Time 1 M2 = ChildrenÕs behavior at Time 2 Y1 = QOL at Time 1 Y3 = QOL at Time 3 X1 Y1 a = 0.17  cÕ = -0.09  bÕ = -0.13   40 the survivorÕs quality of life. Following Baron and KennyÕs (1986) criteria for mediation, Models 1 (simple model) and 2 (controlling for the childrenÕs age) both indicated that the use of children significantly and independently impacted survivorsÕ quality of life (criterion 1: X on Y) and childrenÕs behavior (criterion 2: X on M). When controlling for the use of children, childrenÕs behavior had a significant influence on quality of life (criterion 3: M on Y). Finally, the effect of the use of children on survivorsÕ quality of life when controlling for childrenÕs behavior is smaller than the total effect (criterion 4). The effect sizes within these two models indicate support for rejecting the null hypothesis.    The third model, which controlled for the survivorsÕ quality of life at T1, did not support original hypothesis and did not indicate a fully mediated model. The following presents Baron and KennyÕs( 1986) mediation criteria and how each item was supported or not supported within Model 3:  1) There must be a significant total effect of the independent variable (X) on the dependent variable (Y): There was no significant total effect of the independent variable (use of children at T1) on the dependent variable (quality of life at T3) while controlling for the survivorsÕ quality of life at T1. (Not supported) 2) There must be a significant effect of the independent variable (X) on the mediating variable (M): There was a trend toward a significant effect of the independent variable (use of children at T1) on the mediating variable (child behavior at T2) while controlling for survivorsÕ quality of life at T1. (Not supported) 3) The effect of the mediating variable (M) on the dependent variable (Y) must be significant when controlling for the independent variable (X): The effect of the mediating variable (child behavior at T2) on the dependent variable (survivorsÕ quality of life at T3) was not  41 significant when controlling for the independent variable (use of children at T1) and survivorsÕ quality of life at T1. (Not supported) 4) The effect of the independent variable (X) on the dependent variable (Y) when controlling for the mediating variable (M), must be smaller than the total effect of X on Y (as found in number 1): The effect of the independent variable (use of children at T1) on the dependent variable (quality of life at T3) when controlling for the mediating variable (child behavior at T2) was smaller than the total effect of the use of children at T1 on survivorsÕ quality of life at T3. (Supported) As a result, the null hypothesis is accepted for this model. Taken together, the three models indicate mixed results for the mediating model with all three suggesting a relationship between the use of children and childrenÕs behavior.    42 CHAPTER 5: DISCUSSION  Prior research and anecdotal evidence have indicated that use of children by abusers is an additional tactic of coercive control that impacts survivors of intimate partner violence. This study seeks to expand the scant existing literature by examining the impact of the use of children on survivorsÕ well-being. Three prior studies have examined what predicts the use of children as an abuse tactic (Beeble, Bybee & Sullivan) and how the use of children impacts survivorsÕ symptoms of depression and posttraumatic stress disorder (Ahlfs-Dunn & Huth-Bocks, 2016; Rivera, Sullivan, Zeoli, & Bybee, 2016). However, no study to date has examined the use of children and its impact on survivorsÕ quality of life, and how childrenÕs behavior may mediate the relationship between the two.  While the relationship between the use of children and quality of life was not fully mediated by childrenÕs behavior across the three models, there was still evidence that higher levels of the abuser using children as an abuse tactic negatively impacted survivorsÕ quality of life and childrenÕs behavior. The third model, which included controlling for the survivorsÕ quality of life at T1, suggests that the survivorsÕ quality of life at T1 may be the strongest predictor of their quality of life at T3. As indicated earlier, quality of life at T1 was highly correlated with quality of life at T3 (! = 0.75, p < .001). Due to the strong relationship between quality of life at T1 and quality of life at T3, it is difficult to disentangle the variance within quality of life at T3 as explained by other variables beyond quality of life at T1.  Interestingly, the abuserÕs relationship to the children did not show any significant influence in the models, and ultimately was removed from the final models as a control variable. The study conducted by Beeble and colleagues (2007), which examined the baseline data related to this study, found a relationship between the relationship to the abuser and the use of  43 childrenÑbiological fathers engaged in more tactics related to the use of the children than stepfathers, father figures, and non-father figures. In this study, a correlation was found between the biological fathers and the use of children at T2 (! = 0.24, p < .05) yet there was no change within the model when the relationship statuses were included as control variables. This finding highlights that while there may still be a relationship between the abuser-child relationship and the use of children the abuser-child relationship does not have an impact on the consequences of the use of children, including the childrenÕs behavior and the survivorÕs quality of life. While this may need further exploration, it underscores that childrenÕs behavior may be influenced by their being used as a tactic of abuse against their mothers, regardless of their relationship to who may be using them. Similarly, the survivorsÕ quality of life may be affected as a result of their children being used an abuse tactic of coercion and control, irrespective of the childrenÕs relationship to the abuser.  There is a well-documented link between other forms of abuse, including psychological, physical, and economic abuse, and their negative impact on survivorsÕ well-being including their quality of life (Adeodato, Carvalho, Siqueira, & Souza, 2005; Beeble, Bybee, Sullivan, & Adams, 2009; Bybee & Sullivan, 2002; Laffaye, Kennedy, & Stein, 2003; Leung et al., 2005; Sullivan, 2003; Sullivan & Bybee, 1999; Bonomi et al., 2006; Dutton, Green, & Kaltman, 2006; Mechanic, Weaver, & Resick, 2008a; Nathanson, Shorey, Tirone, & Rhatigan, 2012; Pico-Alfonso, 2005; Theran, Sullivan, Bogat, & Stewart, 2006). The accounts of how batterers use children as an abuse tactic to psychologically impact survivors have largely been anecdotal and lacking empirical investigation. This study has indicated that there is evidence of the link between the use of children and survivorsÕ quality of life. Combined with the studies conducted by Rivera (2016), Ahlfs-Dunn (2016) and their colleagues, there is an increased understanding  44 about the impact of the use of children, including interference with parenting and abuse of children as a tool in coercive control on survivorsÕ well-being.  When children are part of a family, the survivorsÕ intimate partner violence goes beyond the intimate partner-survivor to include the children who have witnessed directly and/or indirectly their parentÕs abuse. Similar to our understanding of the impact of abuse on the survivor parents, there is substantial evidence of the impact of witnessing abuse on children at various developmental stages (Bauer, 2006; Brockie, Dana-Sacco, Wallen, Wilcox, & Campbell, 2015; Edleson, 1999a; Gewirtz & Edleson, 2007; Holt, Buckle, & Whalen, 2008; Homes, 2012;  Levendosky, Huth-Bocks, & Semel, 2002; Lundy & Grossman, 2005; Martinez-Torteya, Bogat, Von Eye, & Levendosky, 2009; Moore & Pepler, 1998; Moylan et all, 2010; Osofsky, 2003). What has been missing from this literature has been how children may also be used as involuntary participants in the psychological abuse of their survivor parents as well. This study has contributed to the literature by documenting the experience of those children who have been used as tools of the abuse against their survivor parents and establishing a link between the abuserÕs use of them and its impact on the childrenÕs lives. The development of the Use of Children scale (Beeble, Bybee, & Sullivan, 2007) has contributed to a broader definition of the direct and indirect ways children may witness their survivor parent being abused. Linking the use of children to child behavior gives us a better indication of how powerful an impact the use of children can have within a family.      Strengths and Limitations Limitations The results of this study should be considered in light of both its strengths and limitations. Due to the small sample size of this study (n = 105), it was difficult to achieve power  45 to find medium to large effect sizes. As a result, the sample size was not large enough to withstand multiple variables, and a decision had to be made to remove non-significant control variables from the final model. Additionally, the model in this study was affected by multicollinearity in two ways. First, the different forms of abuse have long been determined to be affected by multicollinearity. To account for this in the current study, individual mediation models were run with each control variable included separately. Within the model, psychological abuse was highly correlated with the use of children as well as the childrenÕs behavior four months later. The second issue of multicollinearity was related to the makeup of the time-ordered mediation model. Originally, this model was designed to control for the mediator (child behavior) and the dependent variable (quality of life) at the first time point. While it was anticipated that a construct measured across time points would be highly correlated, the smaller sample size made identifying variation in the model difficult when controlling for child behavior at the first time point. Another limitation related to the creation of the model was the decision to focus on only three time points over the course of eight months. This model only focused on a brief snapshot during the timespan of the relationship for many of the survivors and abusers. Repeated measures of the model variables may provide even more evidence for the use of children as an abuse tactic.  This study was also limited by the insufficient racial and ethnic diversity within the sample. Almost half of the women identified as white or Caucasian (n = 56, 51.9%); one-quarter identified as Black or African American (n = 31; 28.7%) with the remaining participants dispersed across several other categories. Future research should focus on capturing a wider  46 range of experiences by more diverse participants. Such efforts would help inform the continued development of necessary culturally specific services.  Additionally, the scale used to measure the use of the children as an abuse tactic uses items that operationalize the construct broadly, without giving specific examples of the tactics related to each item within the scale. For example, each survivor was asked whether the assailant had used the children to harass or intimidate her, stay in her life, keep track of her, or frighten her; or if the assailant had tried to turn the children against her or had tried to convince the children that she should take them back. While these items were designed to be broad in order to try to capture as many experiences as possible, there were some tactics that may not have been captured without having given survivors specific examples (e.g., abusing the children or threatening the children to control the survivorÕs behavior; or withholding money or resources to directly provide for the children). Similar to the other scales that have been created related to physical, psychological, and emotional abuse, asking about specific behaviors may better capture survivorsÕ concrete experiences.   Finally, the use of children measure does not ask about just one child in the family but asked if the abuser has used any of the tactics in general; we were unable to determine if the target children were those children who were necessarily the ones ÒusedÓ in the use of children scale. While this may confound the findings in the study, it could also be argued that when an abuser uses any children as an abuse tactic, it would have similar impacts as other forms of abuse to which children are exposed. Strengths  While there are a number of limitations related to this study, there are several strengths as well. The survivors who comprised the sample of this study were recruited from the community  47 via help-seeking locations that were typically used at times of crises, which increases our understanding surrounding this specific group of survivors. The use of a longitudinal data set, in combination with the application of a time-ordered mediation model, also contributes to the studyÕs strengths. Using a time-ordered mediation model can support more robust findings, taking into account the idea that time must pass before detecting significant change. Additionally, this is one of the few studies to examine the use of children as a specific construct and adds to our limited understanding of measuring an aspect of abuse that many survivors with children experience. Research, Policy, and Practice Implications Need for Continued Research on the Use of Children This study highlights the need for continued research regarding the use of children as an abuse tactic employed to control and psychologically abuse survivors. When abusers choose to be violent toward their partners and there are children within the family, they are making a parenting choice as well. The measure of the use of children taken together with the index developed by Rivera and colleagues (2016) as well as the measure created by Ahlfs-Dunn and Booth-Hocks (2016) contributes to a broader understanding of how children and parenting decisions can be used as tools for continued coercive control of survivors. With this knowledge, more comprehensive and generalizable studies examining this construct would contribute to the understanding of how we can develop and improve services for survivors of intimate partner violence who have children. Going beyond the impact of the survivorsÕ individual well-being, it is also important to consider how the use of children may influence concrete experiences in survivorsÕ and their childrenÕs lives, including custody outcomes, interactions within the child welfare system, and survivorsÕ abilities to access needed resources.   48 Future research needs to broaden the conceptualization of survivorsÕ well-being within this context as well. We should continue to include survivorsÕ quality of life and symptoms of depression and PTSD within the scope of well-being, but when the initial construct (use of children) involves the act of parenting, the final outcome should also examine parenting in relation to a survivorÕs well-being. While there are a number of scales related to parenting and studies focused on parenting stress, few have been helpful within the context of intimate partner violence as they are typically focused on the stress of the survivor parent as a negative impact on their children. In response to this, it would be more useful and survivor-centered to explore the use of children and its impact on a survivorsÕ parenting confidence and efficacy. In addition to further examining the impact of the use of children on survivorsÕ well-being and ability to parent within the context of an abuserÕs interference, it would be just as important to explore the direct impact on the children themselves. This study highlights the importance of asking about the use of each child within the family rather than asking about the abuserÕs use of children in general. Additionally, the impact on children can be measured from the survivorÕs perspective, but it is just as important to measure the childÕs perspective as well. Interviewing children within the context of intimate partner violence can be difficult at best, considering their relationships (or non-relationships) with the abusers, yet their perspectives need to be heard. Additionally, child outcomes can be measured in a number of ways to triangulate with the survivorÕs frame of reference, including the use of school records and other adultsÕ perceptions of the childÕs behavior. Policy and Practice  While many who work in the gender-based violence movement have a strong understanding of how children are used by abusers as a tactic to control their survivor parent,  49 there are many outside of the movement providing services to survivors who do not understand these dynamics (e.g., child welfare). This study contributes to a broader understanding of how intimate partner violence impacts survivors and families, and how abusers may use children as a tactic in their abuse, which can happen long after a relationship has ended. This knowledge can help shape how our social systems respond when families are engaged both voluntarily and involuntarily.  Within the child welfare system, many families become involved due to charges of Òfailure to protect,Ó in which the system identifies that when a survivor parent is in an abusive relationship, they have failed to protect their children. In addition, when families become involved with child welfare due to an actual child abuse charge, workers may discover the co-occurrence of IPV after one or more interactions with the family. When workers have a broad understanding of domestic violence as a pattern-based behavior that often involves children, they are better able to serve families through accurate documentation and a higher probability of accountability of the abusive partner.  Similarly, the civil legal system and family courts would benefit from this knowledge. Historically, these systems, including the child welfare system, have targeted their focus on mothers, holding them at a much higher standard than fathers, even in circumstances of an abusive father. When confronted with the knowledge and documentation of what it means to use children within the context of IPV and the impact on survivors and children, child welfare workers, attorneys, legal guardians ad litem, and judges can better hold abusers accountable and connect survivors and their children with appropriate and essential services. For organizations that already have a deep understanding of gender-based violence, intimate partner violence, and coercive control, this research can give service providers even  50 more insight into survivorsÕ and their childrenÕs experiences of the abuse. With a better understanding of how children may be used as tools of abuse, service providers can create better intervention strategies to target issues that have arisen as a result of familiesÕ experiences, such as: the broken bond between the children and the survivor parent, children displaying a lack of respect or trust toward the survivor parent, poor survivor-child interactions, and increased isolation within the family. Conclusion Intimate-partner violence continues to be a pervasive problem in the United States. This study was designed to help support further understanding of IPV and its impact on survivors and their families. There is scant empirical evidence of how abusers use children to control and manipulate their partners, including understanding the potential outcomes of this behavior. As one of two known quantitative longitudinal studies looking at the use of children as a predictor, it contributes to the field by providing further evidence about the impact of the use of children on survivorsÕ well-being, including quality of life, as mediated by their childrenÕs behavior.     51            APPENDICES    52 APPENDIX A: Use of Children Scale Some men use the children to control the women they are or have been involved with. Using this card, in the last four months, to what extent, if at all, has (A) ___________use the children to:   NONE A LITTLE SOME VERY MUCH N/A 1. Stay in your lifeÉÉÉÉÉÉÉÉ. 1 2 3 4 8 2. Harass youÉÉÉÉÉÉÉ.ÉÉÉ. 1 2 3 4 8 3. Intimidate youÉÉÉÉÉÉÉ.É... 1 2 3 4 8 4. Keep track of youÉÉÉÉ.ÉÉÉ. 1 2 3 4 8 5. Frighten youÉÉÉÉÉ.ÉÉÉÉ.. 1 2 3 4 8 Has he:      6. Tried to turn the kids against youÉÉ 1 2 3 4 8 7. Tried to convince the kids you should take him backÉÉÉÉÉÉÉÉÉÉ 1 2 3 4 8      53 APPENDIX B: Child Behavior Please circle the number describing how often the behavior currently occurs with your child.  Never 1  2  3 Some-times 4  5  6  7 1. Dawdles in getting dressed 1 2 3 4 5 6 7 2. Dawdles or lingers at mealtime 1 2 3 4 5 6 7 3. Has poor table manners 1 2 3 4 5 6 7 4. Refuses to eat food presented 1 2 3 4 5 6 7 5. Refuses to do chores when asked 1 2 3 4 5 6 7 6. Slow in getting ready for bed 1 2 3 4 5 6 7 7. Refuses to go to bed on time 1 2 3 4 5 6 7 8. Does not obey house rules on his/her own 1 2 3 4 5 6 7 9. Refuses to obey until threatened with punishment 1 2 3 4 5 6 7 10 Acts defiant when told to do something 1 2 3 4 5 6 7 11. Argues with parents about rules 1 2 3 4 5 6 7 12. Gets angry when doesnÕt get his/her own way 1 2 3 4 5 6 7 13. Has temper tantrums 1 2 3 4 5 6 7 14. Sasses adults 1 2 3 4 5 6 7 15. Whines 1 2 3 4 5 6 7 16. Cries easily 1 2 3 4 5 6 7 17. Yells or screams 1 2 3 4 5 6 7 18. Hits parents 1 2 3 4 5 6 7 19. Destroys toys and other 1 2 3 4 5 6 7  54 objects 20. Is careless with toys and other objects 1 2 3 4 5 6 7 21. Steals 1 2 3 4 5 6 7 22. Lies 1 2 3 4 5 6 7 23. Teases or provokes other children 1 2 3 4 5 6 7 24. Verbally fights with friends his/her own age 1 2 3 4 5 6 7 25. Verbally fights with sisters and brothers 1 2 3 4 5 6 7 26. Physically fights with friends his/her own age 1 2 3 4 5 6 7 27. Physically fights w/ siblings 1 2 3 4 5 6 7 28. Constantly seeks attention 1 2 3 4 5 6 7 29. Interrupts 1 2 3 4 5 6 7 30. Is easily distracted 1 2 3 4 5 6 7 31. Has short attention span 1 2 3 4 5 6 7 32. Fails to finish tasks or projects 1 2 3 4 5 6 7 33. Has difficulty entertaining himself/herself alone 1 2 3 4 5 6 7 34. Has difficulty concentrating on one thing 1 2 3 4 5 6 7 35. Is overactive or restless 1 2 3 4 5 6 7 36. Wets the bed 1 2 3 4 5 6 7 37. Clings to adults or too dependent 1 2 3 4 5 6 7 38. Complains of loneliness 1 2 3 4 5 6 7 39. Fears she/he might think or do something bad 1 2 3 4 5 6 7 40. Fears or complains that no one 1 2 3 4 5 6 7  55 loves him/her 41. Feels worthless or inferior 1 2 3 4 5 6 7 42. Would rather be alone than with others 1 2 3 4 5 6 7 43. Underactive, slow moving, or lacks energy 1 2 3 4 5 6 7 44. Unhappy, sad, or depressed 1 2 3 4 5 6 7 45. Worried 1 2 3 4 5 6 7    56 APPENDIX C: Quality of Life Scale Now IÕd lake to ask you how you feel about various parts of your life. Please tell me the feelings you have in generalÑtaking into account what has happened in the last four months.  After I ask you each question, please tell me how you feel about that part of your life: either ÒEXTREMELY PLEASED,Ó PLEASED,Ó ÒMOSTLY SATISFIED,Ó ÒEQUALLY DISSATISFIED AND SATISFIED,Ó ÒMOSTLY DISSATISFIED,Ó ÒUNHAPPY,Ó or ÒTERRIBLEÓ. If you feel like a question doesnÕt apply to you, just tell me.  EXTREMELY PLEASEDÉÉÉÉÉÉÉÉÉ. 1 PLEASEDÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ... 2 MOSTLY SATISFIEDÉÉÉÉÉÉÉÉÉÉ... 3 EQUALLY DISSATISFIED AND SATISFIED... 4 MOSTLY DISSATISFIEDÉÉÉÉÉÉÉÉÉ 5 UNHAPPYÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ. 6 TERRIBLEÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ. 7 (Not Applicable) ÉÉÉÉÉÉÉÉÉÉÉÉÉ 8   1. First, a very general question. How do you feel about your life overall?..................... _____ 2. How do you feel about yourself?.................................................................................. _____ 3. How do you feel about your personal safety?............................................................... _____ 4. How do you feel about the amount of fun and enjoyment you have?.......................... _____ 5. How do you feel about the responsibilities you have for members of your family?.... _____ 6. How do you feel about what you are accomplishing in your life?............................... _____ 7. How do you feel about your independence or freedom Ð that is, how free you feel to live the kind of life you want? ......................................................................................... _____ 8. How do you feel about your emotional and psychological well-being?....................... _____ 9. How do you feel about the way you spend your spare time?....................................... _____     57            REFERENCES     58 REFERENCES  Achenbach, T., & Edelbrock, C. (1983). Manual for the child behavior profile and child behavior checklist. Burlington, VT: Author.  Adams, A. E., Sullivan, C. M., Bybee, D., & Greeson, M. R. (2008). Development of the scale of economic abuse. Violence Against Women, 14(5), 563-588.  Adeodato, V. G., Carvalho, R. d. R., Siqueira, V. R. d., & Souza, F. G. d. M. (2005). Quality of life and depression in women abused by their partners. Revista de saude publica, 39(1), 108-113.  Aguilar, R. J., & Nightingale, N. N. (1994). The impact of specific battering experiences on the self-esteem of abused women. Journal of Family Violence, 9(1), 35-45.  Ahlfs-Dunn, S. M., & Huth-Bocks, A. C. (2016). Intimate Partner Violence Involving Children and the Parenting Role: Associations with Maternal Outcomes. Journal of Family Violence, 31(3), 387-399.  Anderson, M. A., Gillig, P. M., Sitaker, M., McCloskey, K., Malloy, K., & Grigsby, N. (2003). 'Why Doesn't She Just Leave?": A descriptive study of victim reported impediments to her safety. Journal of Family Violence, 18(3), 151-155.  Andrews, F. M., & Withey, S. B. (1976). Social indicators of well-being: AmericansÕ perception of quality of life. The Management Group.  Arendell, T. (1992). The social self as gendered: A masculinist discourse of divorce. Symbolic Interaction, 15(2), 151-181.  Bancroft, L. (2003). Why does he do that?: Inside the minds of angry and controlling men. New York: Berkley Books. Bancroft, L., & Silverman, J. G. (2002). The batterer as parent: Addresssing the impact of domestic violence on family dynamics. Newbury Park, CA: SAGE. Bancroft, L., Silverman, J. G., & Ritchie, D. (2011). The batterer as parent: Addressing the impact of domestic violence on family dynamics: Sage Publications. Baron, R. M., & Kenny, D. A. (1986). The moderatorÐmediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of personality and social psychology, 51(6), 1173.  Bauer, N. S., Herrenkohl, T. I., Lozano, P., Rivara, F. P., Hill, K. G., & Hawkins, J. D. (2006). Childhood bullying involvement and exposure to intimate partner violence. Pediatrics, 118(2), e235-e242.   59 Becker, K. D., Stuewig, J., & McCloskey, L. A. (2010). Traumatic stress symptoms of women exposed to different forms of childhood victimization and intimate partner violence. Journal of Interpersonal Violence, 25(9), 1699-1715.  Beeble, M., Bybee, D., & Sullivan, C. (2007). Abusive men's use of children to control their partners and ex-partners. European Psychologist, 12(1), 54-61.  Beeble, M. L., Bybee, D., Sullivan, C. M., & Adams, A. (2009). Main, mediating, and moderating effects of social support on the well-being of survivors of intimate partner violence across 2 years. Journal of Consulting and Clinical Psychology, 77(4), 718-729.  Beeman, S. K., & Edleson, J. L. (2000). Collaborating on family safety: Challenges for children's and women's advocates. Journal of Aggression, Maltreatment & Trauma, 3(1), 345-358.  Bemiller, M. (2008). When Battered Mothers Lose Custody: A Qualitative Study of Abuse at Home and in the Courts. Journal of Child Custody, 5(3), 228-255.  Beydoun, H. A., Beydoun, M. A., Kaufman, J. S., Lo, B., & Zonderman, A. B. (2012). Intimate partner violence against adult women and its association with major depressive disorder, depressive symptoms and postpartum depression: a systematic review and meta-analysis. Social Science & Medicine, 75(6), 959-975.  Black, M. C., Basile, K. C., Breiding, M. J., Smith, S. G., MWalters, M. L., Merrick, M. T., . . . Stevens, M. R. (2011). The National Intimate Partner and Sexual Violence Survey: 2010 Summary Report.  Atlanta, Georgia. Bonomi, A. E., Thompson, R. S., Anderson, M., Reid, R. J., Carrell, D., Dimer, J. A., & Rivara, F. P. (2006). Intimate partner violence and womenÕs physical, mental, and social functioning. American journal of preventive medicine, 30(6), 458-466.  Breiding, M. J., Black, M. C., & Ryan, G. W. (2008). Chronic disease and health risk behaviors associated with intimate partner violenceÑ18 US states/territories, 2005. Annals of epidemiology, 18(7), 538-544.  Brewster, M. (2003). Power and control dynamics in prestalking and stalking situations. Journal of Family Violence, 18(4), 207-217.  Brockie, T. N., Dana-Sacco, G., Wallen, G. R., Wilcox, H. C., & Campbell, J. C. (2015). The Relationship of Adverse Childhood Experiences to PTSD, Depression, PolyAòDrug Use and Suicide Attempt in ReservationAòBased Native American Adolescents and Young Adults. American Journal of Community Psychology, 55(3-4), 411-421.  Browne, A., & Williams, K. R. (1993). Gender, intimacy, and lethal violence: Trends from 1976 through 1987. Gender & Society, 7(1), 78-98.  Bybee, D. I., & Sullivan, C. M. (2002). The process through which an advocacy intervention resulted in positive change for battered women over time. American Journal of Community Psychology, 30(1), 103-132.   60 Campbell, J. C. (2002). Health consequences of intimate partner violence. Lancet, 359(9314), 1331-1336.  Carter, J., & Schechter, S. (1997). Child abuse and domestic violence: Creating community partnerships for safe families: Suggested components of an effective child welfare response to domestic violence San Francisco: Family Violence Prevention Fund. Coker, A. L., Smith, P. H., Bethea, L., King, M. R., & McKeown, R. E. (2000). Physical health consequences of physical and psychological intimate partner violence. Arch Fam Med, 9(5), 451-457.  Collins, L. M., Graham, J. J., & Flaherty, B. P. (1998). An alternative framework for defining mediation. Multivariate Behavioral Research, 33(2), 295-312.  Davies, J., & Lyon, E. (2013). Domestic Violence Advocacy (Vol. 7): Sage. Davies, J. M., Lyon, E., & Monti-Catania, D. (1998). Safety planning with battered women: Complex lives/difficult choices (Vol. 7): Sage Publications, Inc. DeBoard-Lucas, R. L., & Grych, J. H. (2011). The effects of intimate partner violence on school-aged chidlren. In S. A. Graham-Bermann & A. A. Levendosky (Eds.), How intimate partner violence affects chidlren: Developmental research, case studies, and evidence-based intervention. Washington, DC: American Psychological Association. DeKeseredy, W., & Schwartz, M. (2009). Dangerous exits: Escaping abusive relationships in rural America: Rutgers University Press. Dick, G. (2006). Witnessing marital violence as children: men's perceptions of their fathers. Journal of Social Service Research, 32(2), 1-24.  Dutton, M. A., Green, B. L., & Kaltman, S. I. (2006). Intimate partner violence, PTSD, and adverse health outcomes. Journal of Interpersonal Violence, 21(7), 955-968.  Edleson, J. L. (1999a). Children's witnessing of adult domestic violence. Journal of Interpersonal Violence, 14(8), 839-870.  Edleson, J. L. (1999b). The overlap between child maltreatment and woman battering. Violence Against Women, 5(2), 134-154.  Edleson, J. L., & Williams, O. J. (2007). Introduction: Involving men who batter in their childrenÕs lives. Parenting by men who batter, 3-18.  Estefan, L. F., Coulter, M. L., & VandeWeerd, C. (2016). Depression in Women Who Have Left Violent Relationships The Unique Impact of Frequent Emotional Abuse. Violence Against Women, 1077801215624792.  Eyberg, S. M., & Ross, A. W. (1978). Assessment of child behavior problems: The validation of a new inventory. Journal of Clinical Child & Adolescent Psychology, 7(2), 113-116.   61 Finkelhor, D., Turner, H., Ormrod, R., & Hamby, S. L. (2009). Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics, 124(5), 1411-1423.  Fleury, R., E., Sullivan, C. M., & Bybee, D. I. (2000). When ending the relationship does not end the violence: Women's experiences of violence by former partners.  Follingstad, D. R., Rutledge, L. L., Berg, B. J., Hause, E. S., & Polek, D. S. (1990). The role of emotional abuse in physically abusive relationships. Journal of Family Violence, 5(2), 107-120.  Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological science, 18(3), 233-239.  Gewirtz, A. H., & Edleson, J. L. (2007). Young childrenÕs exposure to intimate partner violence: Towards a developmental risk and resilience framework for research and intervention. Journal of Family Violence, 22(3), 151-163.  Golding, J. M. (1999). Intimate partner violence as a risk factor for mental disorders: A meta-analysis. Journal of Family Violence, 14(2), 99-132.  Goodman, L. A., Koss, M. P., Fitzgerald, L. F., Russo, N. F., & Keita, G. P. (1993). Male violence against women: Current research and future directions. American Psychologist, 48(10), 1054.  Graham-Bermann, S. A., & Edleson, J. L. (Eds.). (2001). Domestic Violence in the Lives of Children: The Future of Research, Intervention, ad Social Policy. Washington, D.C.: American Psychological Association. Greif, G. L., & Hegar, R. L. (1993). International parental abduction and its implications for social work practice: Great Britain to the United States. Children & Society, 7(3), 269-276.  Hamby, S., Finkelhor, D., Turner, H., & Ormrod, R. (2010). The overlap of witnessing partner violence with child maltreatment and other victimizations in a nationally representative survey of youth. Child abuse & neglect, 34(10), 734-741.  Hardesty, J., & Ganong, L. (2006). How women make custody decisions and manage co-parenting with abusive former husbands. Journal of Social and Personal Relationships, 23(4), 543.  Hardesty, J. L. (2002). Separation assault in the context of postdivorce parenting. Violence Against Women, 8(5), 597.  Harper, A. (1998). Development of the World Health Organisation WHOQOL-BREF quality of life assessment. Psychological medicine, 28(3), 551-558.  Harrison, C. (2008). Implacably Hostile or Appropriately Protective? Violence Against Women, 14(4), 381.   62 Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach: Guilford Press. Hayes, B. E. (2012). Abusive menÕs indirect control of their partner during the process of separation. Journal of Family Violence, 27(4), 333-344.  Hayes, B. E. (2015). Indirect abuse involving children during the separation process. Journal of Interpersonal Violence, 0886260515596533.  Holden, G. W., Barker, E. D., Appel, A. E., & Hazlewood, L. (2010). Partner-abusers as fathers: Testing hypotheses about their child rearing and the risk of physical child abuse. Partner Abuse, 1(2), 186-199.  Holmes, M. R. (2012). Long-Term Effects of Intimate Partner Violence on Children's Social Behavior. (Dissertation), University of California, Los Angeles.    Holt, S. (2015). PostAòseparation fathering and domestic abuse: Challenges and contradictions. Child Abuse Review, 24(3), 210-222.  Holt, S., Buckley, H., & Whelan, S. (2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child abuse & neglect, 32(8), 797-810.  Jaffe, P. G., Johnston, J. R., Crooks, C. V., & Bala, N. (2008). Custody disputes involving allegations of domestic violence: Toward a differentiated approach to parenting plans. Family Court Review, 46(3), 500-522.  Johnston, J. R., & Girdner, L. K. (2001). Family abductors: descriptive profiles and preventive interventions: US Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention Washington, DC. 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(2), 339.  Kurz, D. (1996). Separation, divorce, and woman abuse. Violence Against Women, 2(1), 63-81.  Kwako, L. E., Glass, N., Campbell, J., Melvin, K. C., Barr, T., & Gill, J. M. (2011). Traumatic brain injury in intimate partner violence: A critical review of outcomes and mechanisms. Trauma, Violence, & Abuse, 12(3), 115-126.  Laffaye, C., Kennedy, C., & Stein, M. B. (2003). Post-traumatic stress disorder and health-related quality of life in female victims of intimate partner violence. Violence and Victims, 18(2), 227-238.  Leung, K.-f., Liu, F.-b., Zhao, L., Fang, J.-q., Chan, K., & Lin, L.-z. (2005). Development and validation of the Chinese Quality of Life Instrument. Health and quality of life outcomes, 3(1), 1.   63 Levendosky, A. A., Huth-Bocks, A., & Semel, M. A. (2002). Adolescent peer relationships and mental health functioning in families with domestic violence. Journal of Clinical Child and Adolescent Psychology, 31(2), 206-218.  Liss, M. B., & Stahly, G. B. (1993). Domestic violence and child custody. Thousand Oaks, CA: Sage Publications. Lundy, M., & Grossman, S. F. (2005). The mental health and service needs of young children exposed to domestic violence: Supportive data. Families in Society: The Journal of Contemporary Social Services, 86(1), 17-29.  MacMillan, H. L., Wathen, C. N., Jamieson, E., Boyle, M. H., Shannon, H. S., Ford-Gilboe, M., . . . Campbell, J. C. (2009). Screening for intimate partner violence in health care settings: a randomized trial. Jama, 302(5), 493-501.  Maddox, L. M. (2015). Young AdultsÕ Perception of Fathering and the Father-Child Relationship in the Context of Marital Violence. Auburn University.    Mahoney, M. R. (1991). Legal images of battered women: Redefining the issue of separation. Michigan Law Review, 1-94.  Martin, S. L., Rentz, E. D., Chan, R. L., Givens, J., Sanford, C. P., Kupper, L. L., . . . Macy, R. J. (2008). Physical and sexual violence among North Carolina women: Associations with physical health, mental health, and functional impairment. Women's Health Issues, 18(2), 130-140.  MartinezAòTorteya, C., Anne Bogat, G., Von Eye, A., & Levendosky, A. A. (2009). Resilience among children exposed to domestic violence: The role of risk and protective factors. Child Development, 80(2), 562-577.  Mbilinyi, L., Edleson, J., Hagemeister, A., & Beeman, S. (2007). What happens to children when their mothers are battered? Results from a four city anonymous telephone survey. Journal of Family Violence, 22(5), 309-317.  McCloskey, L. A. (2001). The" Medea Complex" among men: The instrumental abuse of children to injure wives. Violence and Victims, 16(1), 19-37.  McDonald, R., Jouriles, E. N., Ramisetty-Mikler, S., Caetano, R., & Green, C. E. (2006). Estimating the number of American children living in partner-violent families. Journal of Family Psychology, 20(1), 137.  McFarlane, J., Malecha, A., Watson, K., Gist, J., Batten, E., Hall, I., & Smith, S. (2005). Intimate partner sexual assault against women: frequency, health consequences, and treatment outcomes. Obstetrics & Gynecology, 105(1), 99.  McGee, R. A., Wolfe, D. A., & Wilson, S. K. (1997). Multiple maltreatment experiences and adolescent behavior problems: Adolescents' perspectives. Development and Psychopathology, 9(01), 131-149.   64 McMahon, M., & Pence, E. (1995). Doing more harm than good? Some cautions on visitation centers. In E. Peled, P. G. Jaffe & J. L. Edleson (Eds.), Ending the Cycle of Violence: Comunity Responses to Children of Battered Women (pp. 186-206). Thousand Oaks: Sage. Mechanic, M. B., Weaver, T. L., & Resick, P. A. (2008a). Mental health consequences of intimate partner abuse a multidimensional assessment of four different forms of abuse. Violence Against Women, 14(6), 634-654.  Mechanic, M. B., Weaver, T. L., & Resick, P. A. (2008b). Risk Factors for Physical Injury Among Help-Seeking Battered Women An Exploration of Multiple Abuse Dimensions. Violence Against Women, 14(10), 1148-1165.  Messing, J. T., Thaller, J., & Bagwell, M. (2014). Factors related to sexual abuse and forced sex in a sample of women experiencing police-involved intimate partner violence. Health & social work, 39(3), 181-191.  Moe, A. M. (2009). Battered women, children, and the end of abusive relationships. Affilia, 24(3), 244-256.  Moe, A. M., & Bell, M. P. (2004). Abject Economics The Effects of Battering and Violence on WomenÕs Work and Employability. Violence Against Women, 10(1), 29-55.  Moore, T. E., & Pepler, D. J. (1998). Correlates of adjustment in children at risk.  Moylan, C. A., Herrenkohl, T. I., Sousa, C., Tajima, E. A., Herrenkohl, R. C., & Russo, M. J. (2010). The effects of child abuse and exposure to domestic violence on adolescent internalizing and externalizing behavior problems. Journal of Family Violence, 25(1), 53-63.  Nathanson, A. M., Shorey, R. C., Tirone, V., & Rhatigan, D. L. (2012). The prevalence of mental health disorders in a community sample of female victims of intimate partner violence. Partner Abuse, 3(1), 59-75.  Nemeth, J. M., Bonomi, A. E., Lee, M. A., & Ludwin, J. M. (2012). Sexual infidelity as trigger for intimate partner violence. Journal of Women's Health, 21(9), 942-949.  Orava, T. A., McLeod, P. J., & Sharpe, D. (1996). Perceptions of control, depressive symptomatology, and self-esteem of women in transition from abusive relationships. Journal of Family Violence, 11(2), 167-186.  Osofsky, J. D. (2003). Prevalence of children's exposure to domestic violence and child maltreatment: Implications for prevention and intervention. Clinical Child and Family Psychology Review, 6(3), 161-170.  Pence, E., & Paymar, M. (1993). Education groups for men who batter: The Duluth model: Springer Publishing Company.  65 Pico-Alfonso, M. A. (2005). Psychological intimate partner violence: The major predictor of posttraumatic stress disorder in abused women. Neuroscience & Biobehavioral Reviews, 29(1), 181-193.  Preacher, K. J., & Selig, J. P. (2012). Advantages of Monte Carlo confidence intervals for indirect effects. Communication Methods and Measures, 6(2), 77-98.  Programs, D. A. I. (2013). Post Separation Power and Control Wheel. In D. A. I. Programs (Ed.). Duluth, MN. Ptacek, J. (1997). The tactics and strategies of men who batter: Testimony from women seeking restraining orders. In A. P. Cardarelli (Ed.), Violence Between Intimate Partners. Needham Heights, MA: Allyn & Bacon. Riger, S., Ahrens, C., Blickenstaff, A., O'Leary, K., & Maiuro, R. (2001). Psychological Abuse in Violent Domestic Relations: New York: Springer. Rivera, E. A., Sullivan, C. M., Zeoli, A. M., & Bybee, D. (2016). A Longitudinal Examination of MothersÕ Depression and PTSD Symptoms as Impacted by Partner-Abusive MenÕs Harm to Their Children. Journal of Interpersonal Violence, 0886260516629391.  Rizo, C. F., Macy, R. J., Ermentrout, D. M., & Johns, N. B. (2011). A review of family interventions for intimate partner violence with a child focus or child component. Aggression and Violent Behavior, 16(2), 144-166.  Robinson, E. A., Eyberg, S. M., & Ross, A. W. (1980). The standardization of an inventory of child conduct problem behaviors. Journal of Clinical Child & Adolescent Psychology, 9(1), 22-28.  Rossman, B. B. R., Hughes, H. M., & Rosenberg, M. S. (2000). Children and interparental violence: The impact of exposure: Routledge. Sackett, L. A., & Saunders, D. G. (1999). The impact of different forms of psychological abuse on battered women. Violence and Victims, 14(1), 105-117.  Saltzman, L., Fanslow, J., McMahon, P., & Shelley, G. (2002). Intimate partner violence surveillance: uniform definitions and recommended data elements: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Shalansky, C., Ericksen, J., & Henderson, A. (1999). Abused women and child custody: the ongoing exposure to abusive exâ•’partners. Journal of Advanced Nursing, 29(2), 416-426.  Slote, K. Y., Cuthbert, C., Mesh, C. J., Driggers, M. G., Bancroft, L., & Silverman, J. G. (2005). Battered mothers speak out. Violence Against Women, 11(11), 1367-1395.  Stark, E. (2007). Coercive control: The entrapment of women in personal life: Oxford University Press, USA.  66 Sullivan, C. M. (2003). Using the ESID model to reduce intimate male violence against women. American Journal of Community Psychology, 32(32), 3-4.  Sullivan, C. M., & Bybee, D. I. (1999). Reducing violence using community-based advocacy for women with abusive partners. Journal of Consulting and Clinical Psychology, 67(1), 43.  Sullivan, C. M., Juras, J., Bybee, D., Nguyen, H., & Allen, N. (2000). How children's adjustment is affected by their relationships to their mothers' abusers. Journal of Interpersonal Violence, 15(6), 587-602.  Sullivan, C. M., Tan, C., Basta, J., Rumptz, M., & Davidson, W. S. (1992). An advocacy intervention program for women with abusive partners: Initial evaluation. American Journal of Community Psychology, 20(3), 309-332.  Sutherland, C. A., Bybee, D. I., & Sullivan, C. M. (2002). Beyond bruises and broken bones: the joint effects of stress and injuries on battered women's health. American Journal of Community Psychology, 30(5), 609-636.  Tan, C., Basta, J., Sullivan, C. M., & Davidson, W. S. (1995). The role of social support in the lives of women exiting domestic violence shelters an experimental study. Journal of Interpersonal Violence, 10(4), 437-451.  Theran, S. A., Sullivan, C. M., Bogat, G. A., & Stewart, C. S. (2006). Abusive Partners and Ex-Partners Understanding the Effects of Relationship to the Abuser on WomenÕs Well-Being. Violence Against Women, 12(10), 950-969.  Tjaden, P., & Thoennes, N. (1998). Prevalence, Incidence, and Consequences of Violence against Women: Findings from the National Violence against Women Survey. Research in Brief.  Tjaden, P., & Thoennes, N. (2000). Prevalence and consequences of male-to-female and female-to-male intimate partner violence as measured by the National Violence Against Women Survey. Violence Against Women, 6(2), 142-160.  Tolman, R. M. (1989). The development of a measure of psychological maltreatment of women by their male partners. Violence and Victims, 4(3), 159-177.  Tubbs, C., Williams, O., & Edelson, J. (2007). Shared parenting after abuse: Women's perspectives on parenting after dissolution of a violent relationship. Parenting by men who batter: New directions for assessment and intervention, 19-44.  Vitanza, S., Vogel, L., & Marshall, L. L. (1995). Distress and symptoms of posttraumatic stress disorder in abused women. Violence and Victims, 10(1), 23-34.  Ware, J., Snow, K., Kosinski, M., & Gandek, B. (1993). SF-36 health survey: manual and interpretive guide. Boston: The Health Institute, New England Medical Center.   67 Wittenberg, E., Joshi, M., Thomas, K. A., & McCloskey, L. A. (2007). Measuring the effect of intimate partner violence on health-related quality of life: A qualitative focus group study. Health and quality of life outcomes, 5(1), 1.  Wuest, J., FordAòGilboe, M., MerrittAòGray, M., Varcoe, C., Lent, B., Wilk, P., & Campbell, J. (2009). AbuseAòRelated Injury and Symptoms of Posttraumatic Stress Disorder as Mechanisms of Chronic Pain in Survivors of Intimate Partner Violence. Pain medicine, 10(4), 739-747.  Wuest, J., Merritt-Gray, M., Ford-Gilboe, M., Lent, B., Varcoe, C., & Campbell, J. C. (2008). Chronic pain in women survivors of intimate partner violence. The Journal of Pain, 9(11), 1049-1057.