Gender Mistrust and Intimate Partner Violence During Adolescence and Young Adulthood

Abstract

Recent research has demonstrated that gender mistrust influences the formation and stability of intimate relationships. However, several scholars have noted that individuals form unions despite high levels of mistrust. Yet limited work has examined the implications of gender mistrust for relationship quality, including the experience of intimate partner violence. Furthermore, few quantitative studies have examined correlates of gender mistrust, particularly with regard to identifying factors associated with changes in feelings of mistrust over time. Using the Toledo Adolescent Relationships Study, we examined longitudinal predictors of gender mistrust among a sample of adolescents and young adults (n = 4,466 observations) and found that although parents’ gender mistrust and sociodemographic characteristics were associated with initial levels of mistrust, individuals’ own relationship experiences further shaped trajectories of gender mistrust over time. Additionally, feelings of mistrust corresponded to heightened odds of intimate partner violence perpetration, and this association appeared especially salient for women.

Keywords

adolescents domestic abuse/violence intimate relationships quantitative life course

Introduction

Family researchers have demonstrated that gender mistrust influences a range of relationship outcomes, including lower marriage rates and relationship instability (e.g., Burton, Cherlin, Winn, Estacion, & Holder-Taylor, 2009; Carlson, McLanahan, & England, 2004; Edin & Kefalas, 2005; Furstenberg, 2001). Most of this work has focused on women, suggesting that their general lack of trust in the opposite sex has led many women to conclude that men are not “worth a lifetime of commitment,” nor can most men “be faithful to only one woman” (Edin, 2000). Concerns about sexual exclusivity are particularly common among violent couples as researchers have found that infidelity is a key domain of contestation and a frequent underlying cause of disputes that escalate to violence (Giordano, Copp, Longmore, & Manning, 2015; Nemeth, Bonomi, Lee, & Ludwin, 2012). Nevertheless, women continue to actively search for intimate partners and to enter dating and cohabiting relationships despite these apparently high levels of mistrust (Burton et al., 2009; Lichter & Qian, 2008; Manning & Smock, 2005). Although less often the focus of empirical investigation, men may also experience gender mistrust due to partner infidelity and other relationship concerns (Manning, Trella, Lyons, & Du Toit, 2010), and such feelings may influence their conduct in relationships or contribute to a sense of wariness about relationships in general.

Thus, findings from both quantitative and qualitative work suggest that issues of trust influence relationship formation patterns; however, less is known about the influence of gender mistrust on the quality or nature of those relationships once formed. Among adolescents, gender mistrust is associated with jealousy and verbal conflict (Nomaguchi, Giordano, Manning, & Longmore, 2011), providing a potential link between gender mistrust and patterns of violent behavior. To date, however, most discussions implicate gender mistrust as an outcome, rather than a potential cause, of exposure to physical violence (e.g., Cherlin, Burton, Hurt, & Purvin, 2004; Edin, 2000; Estacion & Cherlin, 2010). Thus, whether gender mistrust influences the perpetration of violence within the intimate context has yet to be considered.

In the current investigation, we used data from the Toledo Adolescent Relationships Study (TARS) to highlight the complex association between gender mistrust and the perpetration of IPV, focusing specifically on physical forms of violence perpetrated toward a current/most recent intimate partner. Because few studies have investigated the origins or development of gender mistrust, we examined factors associated with initial levels of mistrust (during adolescence), as well as those influencing within-individual changes in gender mistrust over time using linear growth curve models. This allowed us to examine the influence of early formative experiences (parents’ gender mistrust, coercive parenting), and the role of prior relationship experiences. As few studies have considered the implications of gender mistrust with respect to relationship dynamics, we employed nonlinear growth curve techniques to examine whether changes in gender mistrust correspond to changes in IPV perpetration from adolescence to young adulthood. Attention to this period of the life course is particularly important since peaks in IPV experience occur in early adulthood (e.g., Fritz & O’Leary, 2004; Johnson, Giordano, Manning, & Longmore, 2015; Kim, Laurent, Capaldi, & Feingold, 2008). We expected that gender mistrust would be a function of early social learning (i.e., parents’ gender mistrust, coercive parenting) and sociodemographic factors, as well as experiences garnered in prior intimate relationships (i.e., infidelity, trust, IPV, control). Our approach allowed us to assess whether gender mistrust is associated with heightened risk of IPV perpetration as respondents move into adulthood. Finally, we examined whether the effect of gender mistrust on IPV perpetration was similar for men and women.

The TARS data are well-suited for these analyses as they include detailed information about the family backgrounds and sociodemographic characteristics of respondents, as well as questions assessing parents’ gender mistrust. Furthermore, the TARS provide multiple assessments of the respondents’ own feelings of mistrust and experience with IPV in addition to information about the broader relationship context within which IPV occurs. This study moves beyond prior work and provides insights into the development of gender mistrust, and its influence on relationship functioning during the young adult phase of the life course.

Background

Gender Mistrust and Romantic Relationships

Explanations for the decline in marriage have focused on economic considerations, including labor market conditions and the economic constraints faced by men who live in poor neighborhoods (e.g., Blau, Kahn, & Waldfogel, 2000; Edin, 2000; Harknett & Kuperberg, 2012; Lichter, McLaughlin, & Ribar, 2002; Manning & Smock, 2005; Wilson, 1987). Observed aggregate economic and marital trends have demonstrated that increases in unemployment rates have been accompanied by corresponding decreases in marriage (Schaller, 2012). Additionally, there are marked differences in marriage rates by race/ethnicity and educational attainment (Payne, 2014). To learn more about the precarious nature of marriage in low-income settings in particular, scholars have conducted a number of qualitative studies across a broad range of samples (e.g., Burton et al., 2009; Edin, 2000; Edin, England, & Linnenberg, 2003; Furstenberg, 2001; Manning et al., 2010). One conclusion drawn from this body of literature is that a pervasive culture of gender mistrust exists in economically disadvantaged communities, which carries serious implications for how individuals evaluate intimate relationships—particularly marriage.

In general, gender mistrust is characterized by a set of attitudes and beliefs about men and women, and is most often conceptualized in terms of negative views of the opposite gender. Although these negative views often encompass a broad range of domains, most researchers focus on gender mistrust regarding relationship commitment and sexual exclusivity. More specifically, given that prior work on the decline of marriage has emphasized the economic constraints faced by men (e.g., limited employment opportunities), or marriage market considerations at the aggregate-level (e.g., the lack of marriageable men), most studies on marriage attitudes focus on women. Accordingly, examples of gender mistrust in the literature have tended to reflect women’s direct experiences with men, as well as those of their close family members, peers, and acquaintances. These attitudes and beliefs portray men as unreliable, untrustworthy, and immature (Coley, 2002; Furstenberg, 2001). Yet scholars (e.g., Manning et al., 2010) have acknowledged that women may also possess characteristics that make them less than ideal marital companions, many of which align with the same attributes that limit men’s marriageability. Few studies, however, have examined gender mistrust among both men and women.

Despite the accumulating evidence of a link between gender mistrust and relationship formation patterns, there is limited work that directly addresses how gender mistrust may influence relationship functioning. Rather, an underlying theme of much of this research is that women’s mistrust of men leads them to become wary of relationships, driving them to focus their energy and attention on their children and their own financial futures (Edin, 2000). This is only a part of the story, however, and as Edin (2000) and others have suggested, most women still aspire to marry, and hope to find the “right” man, despite high levels of mistrust. Furthermore, several scholars (e.g., Burton et al., 2009; Estacion & Cherlin, 2010) have found that gender mistrust is not associated with involvement in intimate relationships. In their ethnographic study of low-income mothers, Burton et al. (2009) raised the following question: “If generalized distrust is as influential in the decline of marriage as several recent studies have suggested, why do so many women who declare distrust of men enter into so many relationships . . . ” (p. 1108). This suggests that there is some utility in exploring (a) variability in the experience of mistrust based on structural location, family background, and the character of early relationship experiences and (b) consequences of these variations in gender mistrust for the ongoing nature/quality of relationships that have been formed, and more specifically, the perpetration of violence within these relationships. As previous research primarily has examined the odds of forming relationships or movement into marriage, the current research extends previous work by examining dynamics within relationships.

Origins of Gender Mistrust

Based on findings from ethnographic and quantitative studies, including those reviewed above, scholars have concluded that gender mistrust has negative implications for union formation and stability—particularly among economically disadvantaged populations. Yet the literature is less clear on the factors that influence the development of gender mistrust, and furthermore, whether the observed associations between gender mistrust and relationship formation generalize beyond disadvantaged populations. Additionally, scholars have suggested that gender mistrust is not a stable characteristic, but rather feelings of mistrust likely are responsive to an individual’s own life experiences both within and beyond the intimate realm (e.g., Edin & Kefalas, 2005; Nomaguchi et al., 2011). Yet to date, limited work has attempted to capture the dynamic nature of gender mistrust, and to our knowledge, no study has examined this possibility. In the current investigation, we accounted for traditional predictors of gender mistrust, including economic disadvantage, race/ethnicity, and early social learning factors, as well as experiences accrued during adolescence and across the transition to adulthood. Additionally, we examined whether the correlates of gender mistrust are similar for men and women.

Explanations for why gender mistrust may be more widespread in economically disadvantaged areas often are structural in nature. In his theory on the retreat from marriage in urban communities, Wilson (1987) suggested that declines in low- and semiskilled employment and wages resulted in a “lack of marriageable men.” As a result of men’s decreased marriageability due to their inability to provide a stable source of income for their families, women began to move away from marriage entirely (Coley, 2002). Additionally, the economic uncertainty of life in low-income settings, including persistent joblessness, places undue strain on intimate relationships contributing to discord and a lack of confidence in forming stable long-term unions.

Given that the vast majority of research on gender mistrust has been conducted in low-income settings, our knowledge of the subject is overly representative of feelings of mistrust among economically disadvantaged individuals, and of Black men and women in particular. Based on his work in low-income Chicago communities, Wilson (1996) contended that gender mistrust is more prevalent among Black men and women as compared with men and women from other race/ethnic groups. Differences in employment and earnings may provide a partial explanation for race/ethnic differences in mistrust; however, differences in economic factors alone are unlikely to provide a sufficient explanation suggesting the utility of considering other distinguishing features of male–female relationships including issues of commitment, infidelity, and trust (Carlson & Furstenberg, 2006; Edin et al., 2003; Furstenberg, Morgan, Moore, & Peterson, 1987; Patterson, 1998; Treas & Giesen, 2000). Thus, it is critical to move beyond race/ethnicity and economic differences.

Family experiences may also be the basis for the development of gender mistrust. In addition to economic and cultural influences, as well as individuals’ direct experiences, gender mistrust can be transmitted intergenerationally (Coley, 2002). Family dynamics that may influence these interrelated processes include the following: (a) children may observe directly their mother’s relationships with men, including instances of infidelity or abuse and/or (b) mothers may convey negative messages about men to their daughters, including the notion that men are “irresponsible, untrustworthy, and self-serving” or “(in)capable of taking care of women” (Coley, 2002, p. 102). Although scholars typically have focused on the intergenerational transmission of gender mistrust from mother to daughter (e.g., Coley, 2002), these same processes likely influence feelings of mistrust among boys. There is limited work, however, examining gender as well as other potential correlates of mistrust using predictive models. An exception is research by Nomaguchi et al. (2011) who found that in addition to neighborhood poverty rates and parent–child relationship quality, parents’ own gender mistrust influenced their adolescent children’s reports of gender mistrust. Although these findings shed light on adolescents’ developing feelings of gender mistrust, potential factors that may continue to shape perceptions of mistrust as individuals make the transition to adulthood, and are involved in more serious relationships, were not examined.

In addition to the influence of early formative experiences in the family of origin on the development of mistrust, individuals are further affected by their own experiences—both within the intimate context and beyond. Some individuals, for example, may exhibit a more global sense of mistrust resulting from personal and community disadvantage (C. Ross, Mirowsky, & Pribesh, 2001). Yet this general mistrust also pervades intimate relationships as mistrusting individuals prefer to “keep their guard up” and distance themselves from others, which may include placing little trust in intimate partners. Additionally, infidelity has been identified as a key reason for mistrusting others (Edin, 2000), and issues of fidelity are central concerns for both men and women (Manning et al., 2010). Finally, scholars (e.g., Burton et al., 2009; Edin, 2000; Edin et al., 2003; Leone, Johnson, Cohan, & Lloyd, 2004; Manning et al., 2010) have identified experiences of prior abuse as a basis for, and as a factor that may exacerbate, mistrust in intimate relationships. Thus, although early relationships provide opportunities for relationship-skill building (e.g., Furman, Brown, & Feiring, 1999), exposure to violence or other negative relationship features may influence an individual’s capacity to trust in future relationships.

The studies reviewed above highlight several of the key correlates of gender mistrust identified in the literature. Yet this roster, while providing a descriptive portrait of key influences of gender mistrust, overlooks the extent to which these forces are interrelated. Although it may be useful to examine economic, cultural, and interpersonal sources of mistrust separately in some instances, such an examination overlooks the multifaceted nature of gender mistrust in which these features overlap and become “fused into a system of beliefs” that is embedded in certain contexts and significantly influences how men and women interact with one another (Furstenberg, 2001, p. 237), including their perpetration of intimate partner violence (IPV).

Gender Mistrust and Involvement in Violent Relationships

Issues of jealousy and mistrust are referenced frequently in the IPV literature, and a number of scholars have demonstrated an association between the experience of abuse and gender mistrust. Most of this research, while concluding that past exposure to abuse limits an individual’s ability to trust in future relationships, neglects to consider whether the association between trust and abuse may be reciprocal in nature. Prior research has identified jealousy as an important correlate of IPV, and a key motivation for IPV perpetration (Giordano, Soto, Manning, & Longmore, 2010; Hettrich & O’Leary, 2007; Kerr & Capaldi, 2011). Feelings of jealousy within the intimate context, however, often are linked to issues of fidelity (Manning et al., 2010), and sexual nonexclusivity is a critical source of conflict. Whereas traditional approaches to IPV have viewed men’s jealousy as a risk factor for women’s victimization, recent evidence has suggested that women’s jealousy may also lead to relationship discord (J. Ross, 2011). Although concerns about infidelity may be linked to actual relationship experiences—both past and present—such considerations may also reflect a general mistrust of the opposite sex. Yet in the IPV literature, the association between gender mistrust and IPV remains underdeveloped.

Relying on samples of both adolescents and adults, researchers (e.g., Cascardi & Vivian, 1995; Miller & White, 2003; Nemeth et al., 2012) have indicated that jealousy and trust represent key themes in men and women’s accounts of partner violence. The link between jealousy and IPV is well-documented in the psychological literature as researchers have compared attachment styles, dependency, and jealousy of violent and nonviolent married men (see Holtzworth-Munroe, Stuart, & Hutchinson, 1997), concluding that violent men are more concerned about potential abandonment, more dependent on their wives, and express more jealousy and less trust in their intimate partners. Such depictions characterize jealousy and trust as individual predispositions or traits, giving limited weight to concerns rooted in the relationship itself. Another line of reasoning has focused on men’s hostility toward and objectification of women, which also accommodates the idea of mistrust, but locates the source in societal level socialization and masculine value systems (Dobash & Dobash, 1979).

Conversely, scholars who place greater emphasis on the relationship context have argued that negative views of the opposite sex may influence levels of trust within the context of a current relationship, including concerns about commitment or fidelity—both of which have been identified as precursors to relationship violence (Miller & White, 2003). These views are consistent with research that shows greater variability in IPV across relationships than would be expected based on either the idea of a stable personality trait or the emphasis on violence as an extension of masculinity training (Capaldi & Kim, 2007).

Thus, considerable work has foregrounded the association between gender mistrust and relationship violence, however, whether mistrust influences the perpetration of violence has yet to be considered empirically. We expect variations in gender mistrust to be associated with IPV risk such that higher levels of mistrust will parallel heightened odds of perpetration. Based on prior research and theorizing, it is less clear whether the effect of mistrust on IPV will be similar for men and women. On the one hand, qualitative research on the transmission of attitudes toward the opposite sex indicates that the endorsement of gender mistrust may be more widespread among women. Similarly, researchers have indicated that men are more likely to be unfaithful to their partners, and thus women may be more mistrusting as a direct result of their own experiences. Despite these potential differences in feelings of mistrust between men and women, there is little basis on which to expect gender differences in the effect of mistrust on IPV perpetration. Nevertheless, gender is central to theorizing on partner violence, and consequently we assess the role of gender in the link between mistrust and violence.

Current Study

The current investigation focused directly on variability in gender mistrust, relying on a longitudinal study of the adolescent and young adult experiences of a large heterogeneous sample of respondents. Using predictive models, we first examined correlates of gender mistrust to further understand sociodemographic and family background influences, as well as the role of prior relationship experiences in shaping within-individual changes in levels of gender mistrust over time. Next, we examined whether changes in gender mistrust correspond to changes in IPV perpetration over time, net of a range of time-varying and time-invariant factors—including well-documented risk factors. Here, we focused on variations in self-reported IPV perpetration rather than victimization, as our primary interest was in the extent to which individuals’ feelings of gender mistrust were associated with their own use of violence within the relationship. An important question remains whether a similar set of predictors are associated with gender mistrust among same-sex couples, and furthermore, whether the link between mistrust and IPV is similar for same- and different-sex couples. However, data limitations preclude meaningful analyses of respondents involved in same-sex relationships. Accordingly, we focus exclusively on involvement in heterosexual relationships.

The existing research on gender mistrust may lead to the conclusion that mistrust is a ubiquitous feature of contemporary relationships. Yet prior research on gender mistrust has focused on contexts of disadvantage, and thus relatively little work has considered the correlates of gender mistrust in a more diverse sample. Furthermore, there is limited empirical research examining factors associated with variation in gender mistrust, including family influences and prior relationship experiences, and the existing work is largely cross-sectional. Based on a range of qualitative studies, it seems plausible that relationship experiences at earlier stages of the life course may influence attitudes about the opposite sex, and that those attitudes are then carried forward into subsequent relationships as a form of “relationship baggage.” Research on the developmental significance of intimate relationships reinforces this idea, as prior relationship experiences have been found to influence behavior in adult unions (Meier & Allen, 2009; Rauer, Pettit, Lansford, Bates, & Dodge, 2013). In the current investigation, we examined factors associated with variability in gender mistrust, drawing on a diverse sample of individuals interviewed first as adolescents and then subsequently at various points throughout the transition to adulthood. In this way, we were able to assess gender mistrust longitudinally, focusing particular attention on the ways in which prior relationship experiences influenced individual trajectories of mistrust. We hypothesized that:

Hypothesis 1: In addition to sociodemographic and family background influences, prior negative relationship experiences (i.e., IPV victimization, partner nonexclusivity, partner influence attempts, and partner-specific mistrust) will exert a positive influence on levels of gender mistrust across the study period.

Next, we assessed the role of this variability as an influence on conduct within intimate relationships. Much of what is known about gender mistrust comes from the marital literature and its focus on mistrust as an influential force in the decline of marriage. This focus on union formation, however, neglects to consider potential ways in which gender mistrust influences the dynamics or “inner workings” of nonmarital as well as marital unions. A remaining question, then, is whether gender mistrust influences the quality of those unions formed—an issue that has received inadequate attention—and specifically, whether it is associated with the use of violence in the intimate context. Research on adolescents’ romantic relationships has found that gender mistrust was associated with other negative qualities of adolescents’ relationships, including jealousy and verbal conflict (Nomaguchi et al., 2011), both of which are risk factors for violence. Additionally, researchers examining the structural origins of mistrust have emphasized the extent to which the actions of mistrusting individuals may actually provoke hostile responses (C. Ross et al., 2001). Thus, in the current study, we considered whether changes in gender mistrust were associated with trajectories of IPV perpetration, net of respondents’ own prior relationship experiences as well as a broad range of family background, sociodemographic, and relationship factors. Experiencing coercive parenting in the family of origin is a risk factor for partner violence (e.g., Ehrensaft et al., 2003; Smith, Ireland, Park, Elwyn, & Thornberry, 2011). Additionally, socioeconomic factors, including neighborhood poverty, influence the experience of partner violence (see Capaldi, Knoble, Shortt, Wu, & Kim, 2012). Relationship characteristics, including relationship status, are associated with partner violence such that individuals in cohabiting relationships report higher levels of violence than their counterparts in marital or dating relationships (Brown & Bulanda, 2008). Finally, parenthood increases IPV risk (Vest, Catlin, Chen, & Brownson, 2002). Thus, we have included these variables as controls in our models. We anticipate the following:

Hypothesis 2: Greater gender mistrust is associated with heightened risk of IPV perpetration, accounting for family background, sociodemographic, and relationship factors.

Data and Method

This research drew on longitudinal data from the TARS, which is based on a stratified random sample of 1,321 adolescents and their parents/guardians. The TARS data were collected in the years 2001 (Wave 1), 2002 (Wave 2), 2004 (Wave 3), 2006 (Wave 4), and 2011 (Wave 5). The sampling frame of the TARS study encompassed 62 schools across seven school districts. The initial sample was drawn from enrollment records for 7th, 9th, and 11th grades, but school attendance was not a requirement for inclusion in the study. Respondents participated in structured in-home interviews through a computer-assisted personal interview technique using questionnaires preloaded on laptop computers. Wave 1 also included the administration of a questionnaire to a parent/guardian (typically the mother). Respondents were ages 12 to 19 at Wave 1, 13 to 20 at Wave 2, 15 to 22 at Wave 3, 17 to 24 at Wave 4, and at the time of the most recent interview (Wave 5), 22 to 29 years.

The stratified, random sample was devised by the National Opinion Research Center and included oversamples of Black and Hispanic adolescents. The initial sample included 1,321 respondents and Wave 5 retained 1,021 valid respondents, or 77% of Wave 1. Attrition analyses revealed that subjects retained did not differ significantly on most dimensions, but were somewhat more likely to be female, and to report a stepparent, single-parent, or “other” family structure. Additionally, Black respondents and those reporting low levels of parental education (less than high school) were less likely to be retained. These data are unique because they include indicators of gender mistrust at each interview wave along with measures of IPV. They also include a rich array of background indicators as well as parental reports of gender mistrust.

The current analyses relied on structured interviews conducted at Waves 1 through 5 with a few exclusions including respondents reporting their race as “other” (n = 26). Additionally, the youngest (12 years) and oldest (29 years) observations were dropped as small cell sizes precluded meaningful analyses of these groups (n = 38). The final analytic sample represented an 11-year accelerated cohort design with three overlapping cohorts (ages 13-28 years).

Dependent Variables

Gender mistrust was assessed across all five interviews. At the first three interviews, female respondents were asked about the extent to which they agreed with the following statements: “Guys will say anything to get a girl,” “Most guys are always ‘hitting on’ girls,” and “You can’t trust most guys” (α = .59, .57, and .59 across interviews 1-3). Responses ranged from 1 (strongly disagree) to 5 (strongly agree), and female adolescent gender mistrust was measured as the mean of these items. These questions were not retained across the later waves of the study, and thus gender mistrust at the time of the fourth and fifth interviews was based on responses to the following single item: “You can’t trust most guys.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree). Male adolescents’ mistrust of women was measured as the mean of three items, including “Most girls are too boy crazy,” “Girls will often use a guy to make another guy jealous,” and “You can’t trust most girls around other guys (α = .60, .63, and .62 across interviews 1-3). Responses ranged from 1 (strongly disagree) to 5 (strongly agree), and male adolescent gender mistrust was taken as the mean. Similar to the measure of female gender mistrust, these measures were not available at the fourth and fifth interviews, and thus male gender mistrust is based on responses to the following: “You can’t trust most girls.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree). Gender mistrust was included as a within-subjects factor in the models predicting IPV perpetration.

IPV perpetration was assessed across all five interviews using four items from the Revised Conflict Tactics Scale (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Respondents were asked how often they committed the following acts against their current or most recent partner: “thrown something at him or her”; “pushed, shoved, or grabbed him or her”; “slapped him or her in the face or head with an open hand”; and “hit him or her.” Responses were scored on a 5-point scale that ranged from 1 (never) to 5 (very often). We used a dichotomous measure of relationship violence, distinguishing between those who reported any (1) and no perpetration (0).

Within-Subjects Factors

Prior Relationship Experiences

IPV victimization was based on the same four items from the Revised Conflict Tactics Scale outlined above, but here focused on victimization. We used a dichotomous measure of relationship violence, distinguishing between those who reported any (1) and no victimization (0) at each of the five interviews. Partner nonexclusivity was based on responses to the following question: “How often do you think X has gotten involved with other girls [guys]?” Responses ranged from 1 (never) to 5 (very often). We dichotomized responses to indicate any sexual nonexclusivity. Partner influence attempts was measured as the mean of the following two items: “X sometimes wants to control what I do” and “X always tries to change me.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree). Partner-specific mistrust was based on a single item asking respondents’ level of agreement with the following: “Sometimes X cannot be trusted.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree).

Relationship status assessed whether the current or most recent relationship reported by respondents was a dating, cohabiting, or marital relationship (reference category), or whether the respondent was not in a relationship (single). Parent was a dichotomous variable indicating whether the respondent had any children. Age was measured in years using a continuous variable reported from respondent’s age at the time of the fifth interview.

Between-Subjects Factors

Parent gender mistrust was based on responses to a series of questions assessed during the Wave 1 parent questionnaire following the prompt: “Thinking about teenagers in general, to what extent do you agree or disagree with the following?” Items included “Boys are only after one thing”; “Girls are too aggressive nowadays”; “I think some children have too much freedom to be around the opposite sex”; “Boys and girls play emotional games with each other”; “I think some parents allow their children too much freedom to date”; “It’s better not to get too serious about one boy/girl in high school”; and “Nowadays, girls are too boy crazy” (Nomaguchi et al., 2011). Responses ranged from 1 (strongly disagree) to 5 (strongly agree), and a scale was constructed from the mean across all items. Note that these questions were administered during the Wave 1 parent questionnaire in reference to the parents’ views of adolescent heterosexual relationships. Coercive parenting was measured using a single item from the Wave 1 adolescent report asking respondents: “When you and your parents disagree about things, how often do they push, slap, or hit you?” Responses were dichotomized to indicate any coercive parenting.

We included a series of sociodemographic indicators: gender, race/ethnicity, family structure, and neighborhood poverty. Gender was measured as female (1) and male (0). We included three dichotomous variables to measure race/ethnicity including non-Hispanic White (contrast category), non-Hispanic Black, and Hispanic. Family structure (Wave 1) included the following categories: two biological parents (contrast category), stepfamily, single-parent family, and any “other” family type. Neighborhood poverty was from U.S. census data at the time of the first interview, and indicated the “percentage of population living below the poverty level” in the respondent’s census tract while growing up. Neighborhood poverty was logged in the multivariate analyses to correct for skewness.

Analytic Strategy

We employed both linear and nonlinear growth curve analyses to model gender mistrust and IPV perpetration. In the first set of analyses, gender mistrust was modeled as a function of age, prior relationship experiences, and several controls. We estimated linear mixed-effects models using restricted maximum likelihood estimation in SAS 9.3. For model comparison purposes, we relied on the Akaike and Bayesian information criteria with smaller values indicating better overall model fit. For the second set of analyses, IPV perpetration was estimated using population-averaged logistic regression models for the probability of perpetrating violence toward a current/most recent partner, and was modeled as a function of age, gender mistrust, prior relationship experiences, and controls. All models were examined separately by gender, as the items used to construct the gender mistrust scales were gender specific. Models were estimated using generalized linear mixed models with a logit link in SAS 9.3. In these models, model fit was assessed using the generalized chi-square statistic.

We begin by presenting descriptive statistics for all study variables (Table 1). Next, we estimated an unconditional means model to determine the amount of variation in gender mistrust between- versus within-individuals. We calculated the intraclass correlation coefficient (ICC), ρ, by σ²_u/(σ²_u + σ²_e), which describes the proportion of variation in the outcome variable that lies between individuals. In our sample, the ICC was 0.27 for males and 0.29 for females, indicating that just over a quarter of the variation in gender mistrust is attributable to differences among individuals, whereas the remaining three quarters is within individuals. Next, we estimated an unconditional growth model by entering the linear effect of age. Model 2 added the within-subjects factors, including prior relationship experiences, relationship status, and parental status. Model 3 introduced the between-subjects factors including parent gender mistrust, coercive parenting, and a range of sociodemographic controls. A similar strategy was followed in the second set of analyses predicting IPV perpetration. First, we estimated an unconditional means model to partition the variance component into within- and between-individual variation. Similar to the procedure described above, we computed the ICC, however, in these logistic regression models σ²_e is the variance of the standard logistic distribution (π²/3). The ICC was 0.21 for male respondents and 0.24 for female respondents, indicating that approximately one quarter of the variation in IPV perpetration occurs between individuals, and the remaining three quarters is attributable to within-individual variation. Next, we estimated an unconditional growth model by entering age and age squared to the model. Model 2 introduced the within-subjects factors, and Model 3 added the between-subjects predictors. A final model added the interaction between gender mistrust and female to determine whether the effect of mistrust on IPV perpetration is similar for men and women.

Table 1.

Descriptive Statistics for Study Variables, by Gender (n = 4,466).

	Males (n = 2,126)	Females (n = 2,340)
Outcomes
Gender mistrust	3.16	3.27
IPV perpetration	0.16	0.23
Within-subjects
Age	18.78	18.91
Age squared	367.82	372.92
Prior relationship experiences
IPV victimization	0.32	0.18
Partner nonexclusivity	0.24	0.23
Partner influence attempts	2.14	1.87
Partner cannot be trusted	2.17	2.24
Relationship status (married)
Single	0.18	0.18
Dating	0.70	0.65
Cohabiting	0.09	0.11
Parent	0.11
Between-subjects
Parent gender mistrust	3.56	3.65
Coercive parenting	0.21	0.24
Neighborhood poverty	14.49	15.11
Race/ethnicity (White)
Black	0.26	0.24
Hispanic	0.11	0.12
Family structure (biological parents)
Single parent	0.23	0.24
Stepparent	0.14	0.14
Other family	0.11	0.15
Mother’s education (high school)
Less than high school	0.12	0.13
Some college	0.33	0.33
College or more	0.21	0.22

Note. IPV = intimate partner violence. Adapted from Toledo Adolescent Relationships Study.

Results

Descriptive Results

Descriptive statistics for all study variables (see Table 1) are based on the 4,466 person-periods, or observations. The average level of gender mistrust was 3.16 for male respondents and 3.27 for female respondents (range = 1-5). This corresponds to a response of “neither agree nor disagree” on questions referencing trust of the opposite sex. On average, 16% of male respondent observations and 23% of female respondent observations indicated IPV perpetration. About two fifths of respondents (41%) reported IPV perpetration at some point during the study (results not shown). The average age of respondents during the study period was roughly 19 years, and this ranged from 13 to 28 years. Approximately 32% of male observations and 18% of female observations indicated IPV victimization to have occurred, and about a quarter of both male and female observations reported partner nonexclusivity. At the person-level, we find that roughly half experienced prior victimization and three fifths reported partner nonexclusivity during the study period (results not shown). Average levels of partner influence attempts were relatively low, as were partner-specific levels of mistrust, with average scores of 2.14 and 2.17 for male respondents, and 1.87 and 2.24 for female respondents, respectively, across the five survey waves. At the between-subjects level, parent gender mistrust was 3.56 for male respondents and 3.65 for female respondents, indicating fairly high levels of mistrust (tending toward endorsement of gender mistrust attitudes). Additionally, roughly one fifth of male respondents, and one quarter of female respondents, reported experiencing coercive parenting.

Modeling Gender Mistrust

The first set of analyses addressed how respondents’ prior relationship experiences, in addition to a number of sociodemographic, family background, and relationship characteristics, were related to levels of gender mistrust. Table 2 presented a series of nested growth curve models for gender mistrust among male respondents. Model 1 included only the linear effect of time. Results of this model indicated that the trajectory of gender mistrust is relatively flat, and declines slightly over time. Model 2 added the within-subjects factors. Of the prior relationship experiences, partner nonexclusivity, partner influence attempts, and partner-specific mistrust were positively associated with gender mistrust. This suggests that, net of time, the experience of partner nonexclusivity, partner influence attempts, and partner-specific mistrust correspond to higher levels of gender mistrust among male respondents. Finally, as compared with respondents in marital unions, dating and cohabiting respondents reported greater levels of gender mistrust.

Table 2.

Effects of Relationship Factors, Family Background, and Sociodemographic Characteristics on Trajectories of Gender Mistrust Among Males (n = 2,126).

Fixed effects, composite model	Model 1: Unconditional growth model, b (SE)	Model 2: Linear effect of time and within-subject factors, b (SE)	Model 3: Linear effect of time, within- and between-subjects factors, b (SE)
Within-subjects
Initial status	3.894*** (0.09)	3.297*** (0.18)	2.878*** (0.23)
Age	−0.039*** (0.00)	−0.040*** (0.01)	−0.037*** (0.01)
Prior relationship experiences
IPV victimization		−0.022 (0.04)	−0.048 (0.04)
Partner nonexclusivity		0.210*** (0.04)	0.183*** (0.04)
Partner influence attempts		0.050* (0.02)	0.050* (0.02)
Partner cannot be trusted		0.072*** (0.02)	0.070*** (0.02)
Relationship status (married)
Single		0.131 (0.17)	0.103 (0.17)
Dating		0.310** (0.11)	0.292*** (0.10)
Cohabiting		0.429*** (0.11)	0.407*** (0.11)
Parent		0.197** (0.07)	0.117 (0.07)
Between-subjects
Parent gender mistrust			0.085* (0.04)
Coercive parenting			0.076 (0.05)
Neighborhood poverty			0.003 (0.00)
Race/ethnicity (White)
Black			0.293*** (0.06)
Hispanic			0.208** (0.07)
Family structure (biological parents)
Single parent			0.047 (0.06)
Stepparent			0.070 (0.07)
Other family			0.034 (0.03)
Mother’s education (high school)
Less than high school			0.033 (0.08)
Some college			−0.093 (0.05)
College or more			−0.222*** (0.06)
Variance components
Level 1: Within-person	0.453***	0.454***	0.452***
Level 2: In intercept	1.518***	1.569***	1.604***
Level 2: In rate of change	0.006***	0.005***	0.006***
AIC	6277.2	5150.8	5101.6
BIC	6294.8	5168.3	5119.1

Note. IPV = intimate partner violence; AIC = Akaike information criterion; SE = standard error; BIC = Bayesian information criterion. Adapted from Toledo Adolescent Relationships Study.

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

Model 3 introduced the between-subjects factors, and consistent with prior work, several of the sociodemographic and family background factors were related to gender mistrust. Specifically, male respondents whose parents reported higher levels of mistrust were more mistrusting of the opposite sex. As compared with their White counterparts, Black and Hispanic individuals reported higher levels of mistrust. Mother’s education was negatively associated with gender mistrust; male respondents with college educated parents, compared with those whose parents had a high school degree, reported lower levels of gender mistrust. Net of these factors, however, prior relationship experiences—including partner nonexclusivity, partner influence attempts, and partner-specific mistrust—continued to exert a positive influence on gender mistrust, suggesting that at any given age (of the ages observed in this study) exposure to these different negative features of relationships enhance levels of gender mistrust.

Table 3 included the results of models predicting gender mistrust among female respondents. Similar to Table 2, Model 1 presented the results of the unconditional growth model to assess the trajectory of gender mistrust over time. The results showed that as in the male models, female levels of gender mistrust are fairly stable, with a slight decline over time. The within-subjects factors were included in Model 2, and results indicate that, of the prior relationship experiences, IPV victimization, partner influence attempts, and partner-specific mistrust were positively associated with gender mistrust. In contrast to the models predicting male reports of mistrust, partner nonexclusivity was not related to gender mistrust in these models. Supplemental models examined the effect of each of the prior relationship experiences on gender mistrust individually, and in this more trimmed model, partner nonexclusivity exerted a positive effect on gender mistrust. This effect was attenuated, however, after controlling for the other relationship experiences. Additionally, relationship status was associated with gender mistrust such that single and dating, compared with married, respondents reported higher levels of gender mistrust. Finally, being a parent was positively associated with gender mistrust.

Table 3.

Effects of Relationship Factors, Family Background, and Sociodemographic Characteristics on Trajectories of Gender Mistrust Among Females (n = 2,340).

Fixed effects, composite model	Model 1: Unconditional growth model, b (SE)	Model 2: Linear effect of time and within-subject factors, b (SE)	Model 3: Linear effect of time, within- and between-subjects factors, b (SE)
Within-subjects
Initial status	4.119*** (0.09)	3.165*** (0.16)	2.580*** (0.23)
Age	−0.044*** (0.00)	−0.037*** (0.01)	−0.033*** (0.01)
Prior relationship experiences
IPV victimization		0.120* (0.05)	0.104* (0.05)
Partner nonexclusivity		0.082 (0.04)	0.073 (0.04)
Partner influence attempts		0.070** (0.02)	0.063** (0.02)
Partner cannot be trusted		0.110*** (0.02)	0.104*** (0.02)
Relationship status (married)
Single		0.618** (0.20)	0.589** (0.20)
Dating		0.417*** (0.08)	0.411*** (0.08)
Cohabiting		0.123 (0.08)	0.105 (0.08)
Parent		0.248*** (0.05)	0.173** (0.06)
Between-subjects
Parent gender mistrust			0.124** (0.04)
Coercive parenting			0.077 (0.06)
Neighborhood poverty			0.002 (0.00)
Race/ethnicity (White)
Black			0.150* (0.07)
Hispanic			0.106 (0.08)
Family structure (biological parents)
Single parent			0.081 (0.06)
Stepparent			0.045 (0.07)
Other family			0.096 (0.07)
Mother’s education (high school)
Less than high school			0.121 (0.08)
Some college			−0.036 (0.06)
College or more			−0.125 (0.07)
Variance components
Level 1: Within-person	0.504***	0.495***	0.498***
Level 2: In intercept	1.368***	1.082***	0.951***
Level 2: In rate of change	0.005***	0.004***	0.003***
AIC	7093.0	5788.4	5783.8
BIC	7110.9	5806.2	5801.5

Note. SE = standard error; IPV = intimate partner violence; AIC = Akaike information criterion; BIC = Bayesian information criterion. Adapted from Toledo Adolescent Relationships Study.

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

Model 3 included a broad range of sociodemographic and family background factors measured at the between-subjects level. Similar to the models predicting gender mistrust among male respondents, parent gender mistrust and race were positively associated with gender mistrust among female respondents. That is, higher levels of parent gender mistrust corresponded to higher levels of gender mistrust among female respondents, and Black respondents reported greater levels of gender mistrust than their White counterparts. Controlling for these factors, females’ prior relationship experiences remained significant. Specifically, exposure to IPV victimization, partner influence attempts, and partner-specific mistrust contributed to increases in gender mistrust among female respondents across the observation period.

Modeling IPV Perpetration

The analyses examining IPV perpetration are shown in Tables 4 and 5. As in the results described above, a series of nested models is presented addressing how changes in feelings of gender mistrust influence changes in the odds of IPV perpetration over time. Table 4 included the results of models predicting men’s IPV perpetration. Model 1 presented the results of the unconditional growth model, which included both age and age squared to account for the curvilinear trajectory of IPV perpetration observed over the study period. Shifting attention to the partial slope of age for male respondents, .561 + 2(−0.015)AGE, the odds of IPV perpetration increase until about age 19, and then begin to decline. Model 2 added the within-subjects factors, and net of prior relationship experiences, relationship status, and status as a parent, gender mistrust was not associated with the odds of IPV perpetration among male respondents. Holding relationship status and status as a parent constant, respondents’ prior relationship experiences appeared to have a positive effect on the odds of IPV perpetration. In contrast to the models predicting gender mistrust, however, the effect of IPV victimization was significant and positive in these models. Additionally, partner nonexclusivity and partner influence attempts were associated with greater odds of IPV perpetration. Model 3 included the between-subjects factors. In this model, the effect of gender mistrust remained nonsignificant. Furthermore, none of the between-subjects factors were significantly associated with the odds of IPV perpetration. Controlling for the full range of within- and between-subjects factors, only prior relationship experiences were associated with men’s IPV perpetration. Specifically, men who reported prior IPV victimization, and those who experienced partner nonexclusivity and partner influence attempts, reported greater odds of IPV perpetration across the study period.

Table 4.

Effects of Gender Mistrust, Relationship Factors, Family Background, and Sociodemographic Characteristics on Trajectories of Physical IPV Perpetration Among Males (n = 2,126).

Fixed effects, composite model	Model 1: Unconditional growth model, b (SE)	Model 2: Linear effect of time and within-subject factors, b (SE)	Model 3: Linear effect of time, within- and between-subjects factors, b (SE)
Within-subjects
Initial status	−6.919*** (1.70)	−2.513 (2.08)	−3.339 (2.21)
Age	0.561** (0.17)	−0.145 (0.21)	−0.093 (0.22)
Age squared	−0.015*** (0.00)	0.002 (0.01)	0.000 (0.01)
Gender mistrust		0.095 (0.10)	0.047 (0.10)
Prior relationship experiences
IPV victimization		3.708*** (0.23)	3.693*** (0.24)
Partner nonexclusivity		0.545** (0.18)	0.547** (0.18)
Partner influence attempts		0.258** (0.09)	0.262** (0.09)
Partner cannot be trusted		0.035 (0.08)	0.034 (0.08)
Relationship status (married)
Single		−0.570 (0.75)	−0.705 (0.76)
Dating		−0.691 (0.43)	−0.693 (0.44)
Cohabiting		−0.298 (0.43)	−0.235 (0.44)
Parent		0.306 (0.25)	0.307 (0.26)
Between-subjects
Parent gender mistrust			0.104 (0.16)
Coercive parenting			0.325 (0.20)
Neighborhood poverty			0.001 (0.01)
Race/ethnicity (White)
Black			0.095 (0.24)
Hispanic			−0.089 (0.27)
Family structure (biological parents)
Single parent			0.124 (0.23)
Stepparent			0.117 (0.26)
Other family			0.024 (0.28)
Mother’s education (high school)
Less than high school			−0.308 (0.29)
Some college			0.185 (0.21)
College or more			−0.339 (0.27)
Variance component, intercept
τ	0.914	0.502	0.535

Note. SE = standard error; IPV = intimate partner violence. Adapted from Toledo Adolescent Relationships Study.

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

Table 5.

Effects of Gender Mistrust, Relationship Factors, Family Background, and Sociodemographic Characteristics on Trajectories of Physical IPV Perpetration Among Females (n = 2,340).

Fixed effects, composite model	Model 1: Unconditional growth model, b (SE)	Model 2: Linear effect of time and within-subject factors, b (SE)	Model 3: Linear effect of time, within- and between-subjects factors, b (SE)
Within-subjects
Initial status	−8.371*** (1.46)	−9.291*** (1.96)	−10.403*** (2.09)
Age	0.705*** (0.15)	0.664*** (0.19)	0.718*** (0.20)
Age squared	−0.017*** (0.00)	−0.018*** (0.00)	−0.019*** (0.00)
Gender mistrust		0.220** (0.08)	0.141 (0.08)
Prior relationship experiences
IPV victimization		3.264*** (0.17)	3.299*** (0.18)
Partner nonexclusivity		0.235 (0.17)	0.179 (0.18)
Partner influence attempts		0.283** (0.09)	0.250** (0.09)
Partner cannot be trusted		0.240*** (0.07)	0.232*** (0.07)
Relationship status (married)
Single		−0.382 (0.85)	−0.601 (0.86)
Dating		−0.810* (0.32)	−0.910** (0.32)
Cohabiting		0.287 (0.32)	0.174 (0.32)
Parent		0.249 (0.20)	−0.060 (0.21)
Between-subjects
Parent gender mistrust			0.171 (0.14)
Coercive parenting			0.147 (0.18)
Neighborhood poverty			0.001 (0.01)
Race/ethnicity (White)
Black			0.600* (0.23)
Hispanic			0.347 (0.26)
Family structure (biological parents)
Single parent			0.323 (0.21)
Stepparent			0.081 (0.25)
Other family			0.306 (0.25)
Mother’s education (high school)
Less than high school			0.397 (0.26)
Some college			−0.127 (0.20)
College or more			−0.334 (0.24)
Variance component, intercept
τ	1.133	0.772	0.772

Note. SE = standard error; IPV = intimate partner violence. Adapted from Toledo Adolescent Relationships Study.

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

Turning to the results of the models predicting women’s IPV perpetration, again the trajectory of IPV perpetration over the study period was curvilinear. The partial slope for the effect of age was .705 + 2(−0.017)AGE, suggesting that the odds of IPV perpetration peaks around age 21, and then gradually declines. Model 2 included the within-subjects factors and a number of significant findings emerge. First, net of prior relationship experiences, relationship status, and parental status, gender mistrust was associated with heightened odds of IPV perpetration. Similar to the models predicting gender mistrust, women’s prior relationship experiences exerted a positive effect of the odds of IPV perpetration. Specifically, women’s IPV victimization, partner influence attempts, and partner-specific trust were positively associated with IPV perpetration. Partner nonexclusivity was not significant in this model, however supplemental analyses revealed that this was driven by the addition of the other relationship experiences to the model, and particularly the inclusion of partner-specific mistrust. Additionally, women in dating relationships, as compared with married, reported lower odds of IPV perpetration.

Following the inclusion of the between-subjects factors (Model 3), the effect of gender mistrust was attenuated, and no longer significant at conventional levels (p = .097). The reduction in the magnitude of the coefficient for gender mistrust was due to the addition of sociodemographic factors to the model, particularly poverty, race, and family structure. As in Model 2, women’s prior relationship experiences remained significantly associated with the odds of IPV perpetration in this full model such that IPV victimization, partner influence attempts, and partner-specific mistrust were associated with greater odds of IPV perpetration. Additionally, women in dating relationships, as compared with married, reported lower odds of IPV perpetration, and Black, compared with White, women reported higher odds of IPV perpetration.

A secondary objective of this investigation was to focus particular attention on the role of gender. Because the items comprising the gender mistrust scales differed for male and female respondents, we cannot provide a direct comparison of the models. We can, however, offer a number of observations about the findings, and particularly, how predictors of gender mistrust and IPV perpetration differ across gender. First, the trajectories of gender mistrust were very comparable for male and female respondents; levels of gender mistrust were generally stable with a modest decline over time. Although statistical comparisons cannot be made, examination of reports of gender mistrust for male and female respondents revealed that levels of mistrust were higher among females compared with their male counterparts across study waves. Results of the multivariate models predicting gender mistrust were substantively similar. In particular, prior relationship experiences, relationship status, parent gender mistrust, and race were associated with men’s and women’s reports of gender mistrust. There were some differences, however. For example, whereas IPV victimization was not significantly associated with gender mistrust for men, it was positively associated with gender mistrust for women. Furthermore, partner nonexclusivity was a robust predictor of men’s but not women’s reports of gender mistrust. As we mentioned above, partner nonexclusivity was associated with women’s reports of gender mistrust in more trimmed models, nevertheless, it appeared that the experience of partner nonexclusivity may be more closely associated with feelings of gender mistrust among young men. Involvement in cohabiting and dating, compared with marital, relationships corresponded with higher levels of mistrust among men, whereas being single or dating, compared with married, was positively associated with women’s gender mistrust. Black and Hispanic young men, compared with their White counterparts, reported higher levels of mistrust. Among the young women, Hispanic respondents reported similar levels of gender mistrust as their White peers. Black young women, however, reported higher levels of mistrust. Finally, mother’s education was associated with gender mistrust in the male models, suggesting that young men with college-educated parents reported lower levels of gender mistrust over the course of the study. Mother’s education was not a significant predictor of gender mistrust among young women.

In contrast to the models predicting gender mistrust, results of the models examining the odds of IPV perpetration substantively differed in many respects for young men and women. The trajectories were similar as evidenced by the curvilinear nature of the association between age and IPV perpetration. However, following the addition of the within-subjects factors, gender mistrust was not associated with men’s IPV perpetration, and moreover, the coefficients for age and age squared were no longer significant. Thus, the odds of young men’s IPV perpetration appeared to be most closely tied to their own relationship experiences, including IPV victimization, partner nonexclusivity, and partner influence attempts. Feelings of gender mistrust, and a range of relationship features, on the other hand, influenced young women’s reports of IPV perpetration. These associations largely persisted following the addition of the between-subjects factors; however, controlling for the full roster of study variables reduced the effect of gender mistrust on IPV perpetration to marginal significance.

Furthermore, examination of the generalized chi-square statistic indicated a substantial improvement in model fit from the unconditional growth model (Model 1) following the addition of the within-subjects factors (Model 2) in the models predicting IPV perpetration for men and women (Tables 4 and 5). This is further reflected in the reduction in the age coefficients. In the models focused on women’s IPV perpetration, the introduction of the within-subjects factors in Model 2 slightly reduced the linear and quadratic effects of age (about a 6% reduction), and for men, reductions were more substantial as these coefficients were reduced to nonsignificant levels. These findings suggest that the within-subjects factors, particularly the prior relationship experiences, account for the observed variability in men’s IPV perpetration by age. Although within-subjects factors helped explain women’s variation in IPV perpetration over time, significant variation remained on inclusion of the different relational factors. Supplemental analyses (not shown) examined whether the effect of gender mistrust on men’s IPV perpetration was significant prior to the addition of the full roster of covariates, and controlling for family background and sociodemographic characteristics, gender mistrust was significantly associated with men’s heightened risk of IPV perpetration. After controlling for the more immediate relationship context via the items tapping prior relationship experiences, however, the effect of gender mistrust on perpetration was no longer significant for men.

Discussion

The results of the current analyses suggest that feelings of gender mistrust have implications for the healthy functioning of intimate relationships by heightening the risk of IPV perpetration. Additionally, such feelings appear to be more than simply a consequence of poverty or socioeconomic factors—a conclusion easily drawn based on prior qualitative studies. Drawing on a large, diverse sample of individuals first interviewed as adolescents, and then subsequently at various points across the transition to adulthood, using growth curve analyses, we showed that family influences and prior relationship experiences were significantly associated with gender mistrust. Furthermore, feelings of mistrust led to a heightened risk of IPV perpetration. These findings suggest the need to further examine gender mistrust in relation to the quality of intimate unions, with a particular focus on its role in the resort to violence.

Both qualitative and quantitative studies have forged a link between economic conditions and gender mistrust, and others have acknowledged the potential for intergenerational transmission of attitudes and beliefs about the opposite sex. In the current investigation, in addition to neighborhood poverty, parental education, and race, parent’s own expression of gender mistrust increased respondents’ feelings of gender mistrust. Yet whereas prior research has been largely cross-sectional, the current longitudinal analyses shed additional light on the nature of these associations. That is, although early social learning and other structural considerations are important in the development of feelings of mistrust, they do not appear to alter trajectories of mistrust over time. Thus, individuals learn to be wary of relationships, but those expectations have little opportunity to be reinforced until they become involved with intimate partners (Furstenberg, 2001). Relationship experiences, including partner nonexclusivity, partner control, and partner-specific mistrust, reinforce negative views of the opposite sex and are associated with changes in feelings of gender mistrust over time.

Whereas most of the prior work on gender mistrust focused on relationship formation, our results indicate that feelings of gender mistrust have consequences for the use of violence within the dyadic context. The results indicated that increases in feelings of gender mistrust were associated with women’s increased odds of IPV perpetration at any given point in time across the study period. However, after controlling for a range of time-stable characteristics, the effect of gender mistrust was no longer significant. Supplemental analyses revealed that this was due to the addition of a range of sociodemographic characteristics to the model, including poverty, race, and family structure. A contribution of these findings is that net of individual feelings of trust toward a partner, the more general notion of gender mistrust remained an important factor leading to women’s IPV perpetration. After controlling for the more immediate relationship context, the effect of gender mistrust on men’s IPV perpetration was not significant. These gendered findings warrant further investigation. On the one hand, they point to a more pervasive culture of gender mistrust among women, and this seems largely consistent with prior qualitative research. However, this is incomplete as an explanation. There is a general consensus that feelings of gender mistrust are more widespread among women, but based on this, we should merely expect to see greater endorsement of gender mistrust among women—not a stronger association between such feelings and violence. Nevertheless, these findings call into question the motivations for men’s and women’s perpetration highlighted by typological approaches to gender-based violence, and suggest that the gender dynamics of IPV are much more complex. Additionally, they signal the importance of continuing to complicate and refine our understandings about specific gender-based attitudes that foster these forms of violence.

This investigation is, to our knowledge, the first to examine gender mistrust as a precursor to violence. Furthermore, it is one of a small number of studies to examine gender mistrust beyond contexts of disadvantage, and also, to do so longitudinally. Despite these contributions, this study is not without its own limitations. First, whereas our measure of parent’s gender mistrust is based on seven items tapping several dimensions of gender mistrust, our measure of the respondent’s gender mistrust is more limited. Additionally, it is based on a general measure of mistrust of the opposite sex and thus is not context specific. Prior research has indicated that gender mistrust often stems from sexual infidelity, but likely other considerations influence feelings of mistrust (e.g., concerns about partner’s financial prospects), and some of these considerations may be life course specific. To account for the intergenerational transmission of gender mistrust, we controlled for parent’s gender mistrust. Yet we do not disaggregate parent’s gender mistrust by gender. This approach is consistent with prior work and the notion that gender mistrust “ . . . encompasses people’s general negative views of men and women” (Nomaguchi et al., 2011, p. 1033); however, future work is needed that conceptualizes and measures parent’s gender mistrust differently, such as focusing on the role of gender (of the parent and child) and the parent’s own mistrust based specifically on prior relationship experiences. In addition, other aspects of the respondents’ upbringing, including parents’ unstable relationships, frequent discord, and interparental violence, likely influence feelings of mistrust. Future work may benefit from further attention to the processes underlying transmission of negative views of the opposite sex from one generation to the next. Furthermore, the extent to which gender mistrust is based on heteronormative gender role standards remains unexplored. Future examinations of gender mistrust among same-sex couples may reveal that mistrust takes on unique meanings. Consistent with Umberson’s (1992) notion of gender as a relational concept, for example, women in same-sex couples may experience gender mistrust quite differently than women in different sex couples. The unique ways in which same-sex couples enact gender roles is an issue that merits further attention—especially when focusing on important health behaviors. Although these data offer an unparalleled view of gender mistrust and IPV in a population-based sample, the results are based on respondents from one region of the country. Results should be replicated with nationally representative samples. Our project focuses on young adulthood, a period characterized by peaks in IPV; the observed associations may shift, however, as individuals move into middle age and more stable relationships. Further work that extends the implications of gender mistrust into older age ranges is warranted.

The primary objective of the current investigation was to examine the association between gender mistrust and physical IPV perpetration. However, future work should focus attention on specific ways in which gender mistrust influences the perpetration of relationship violence. Scholars have suggested, for example, that mistrusting individuals may engage in “preemptive actions” that provoke violent responses (C. Ross et al., 2001), including monitoring the whereabouts of significant others, as well as looking through partners’ phones, e-mail, and social media accounts. High levels of gender mistrust may also cause individuals to be mistrusting of a specific partner—particularly regarding issues of sexual exclusivity—and thus fidelity concerns may provide a potential mechanism linking a broader sense of gender mistrust to the use of violence within a specific relationship. Finally, monitoring, fidelity concerns, and other potential consequences of gender mistrust represent behaviors that may elicit verbal discord. This may be particularly true when the monitoring and issues of fidelity are not the result of past behaviors of the partner (i.e., as a result of their own infidelity), but rather stem from one’s own worldview of men/women as untrustworthy. Forging a link to communication processes within the relationship is potentially important, as much prior research (Wilkinson & Hamerschlag, 2005) has documented that high levels of verbal conflict are reliably associated with IPV risk. Whereas gender mistrust is most often examined in relation to union formation, the findings of the current investigation suggest the utility of directing attention to the quality of unions formed. Programs aimed at fostering healthy relationships may benefit from shifting from a singular focus on trust of a specific partner to a more comprehensive view of mistrust and its most proximate causes. Addressing these factors will not only serve to promote healthier relationships but may lessen the probability of IPV.

Footnotes

Authors’ Note

The opinions, findings, and conclusions or recommendations expressed in this publication/program/exhibition are those of the authors and do not necessarily reflect the official views of the Department of Justice or National Institutes of Health.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD036223 and HD044206), the Department of Health and Human Services (5APRPA006009), the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice (Award Nos. 2009-IJ-CX-0503 and 2010-MU-MU-0031), and in part by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R24HD050959).

References

Blau

F. D.

Kahn

L. M.

Waldfogel

(2000). Understanding young women’s marriage decisions: The role of labor and marriage market conditions. Industrial & Labor Relations Review, 53, 624-647.

Brown

S. L.

Bulanda

J. R.

(2008). Relationship violence in young adulthood: A comparison of daters, cohabitors, and marrieds. Social Science Research, 37, 73-87.

Burton

L. M.

Cherlin

Winn

D. M.

Estacion

Holder-Taylor

(2009). The role of trust in low-income mothers’ intimate unions. Journal of Marriage and Family, 71, 1107-1124.

Capaldi

D. M.

Kim

H. K.

(2007). Typological approaches to violence in couples: A critique and alternative conceptual approach. Clinical Psychology Review, 27, 253-265.

Capaldi

D. M.

Knoble

N. B.

Shortt

J. W.

Kim

H. K.

(2012). A systematic review of risk factors for intimate partner violence. Partner Abuse, 3(2), 1-27.

Carlson

McLanahan

England

(2004). Union formation in fragile families. Demography, 41, 237-261.

Carlson

M. J.

Furstenberg

F. F.

(2006). The prevalence and correlates of multipartnered fertility among urban U.S. parents. Journal of Marriage and Family, 68, 718-732.

Cascardi

Vivian

(1995). Context for specific episodes of marital aggression: Gender and severity of violence differences. Journal of Family Violence, 10, 263-281.

Cherlin

A. J.

Burton

L. M.

Hurt

T. R.

Purvin

D. M.

(2004). The influence of physical and sexual abuse on marriage and cohabitation. American Sociological Review, 69, 768-789.

10.

Coley

R. L.

(2002). What mothers teach, what daughters learn: Gender mistrust and self-sufficiency among low-income women. In Booth

Crouter

A. C.

(Eds.), Just living together: Implications of cohabitation on families, children, and social policy (pp. 97-105). Mahwah, NJ: Lawrence Erlbaum.

11.

Dobash

R. E.

Dobash

(1979). Violence against wives: A case against the patriarchy. New York, NY: Free Press.

12.

Edin

(2000). What do low-income single mothers say about marriage? Social Problems, 47, 112-133.

13.

Edin

England

Linnenberg

(2003, September). Love and distrust among unmarried parents. Paper presented at National Poverty Center conference on Marriage and Family Formation Among Low-Income Couples, Washington, DC. Retrieved from http://www.npc.umich.edu/news/events/past/EdinEnglandLinnenberg.pdf

14.

Edin

Kefalas

(2005). Promises I can keep. Berkeley: University of California.

15.

Ehrensaft

Cohen

Brown

Smailes

Chen

Johnson

(2003). Intergenerational transmission of partner violence: A 20-year prospective study. Journal of Consulting and Clinical Psychology, 71, 741-753.

16.

Estacion

Cherlin

(2010). Gender distrust and intimate unions among low-income Hispanic and African American women. Journal of Family Issues, 31, 475-498.

17.

Fritz

P. A.

O’Leary

K. D.

(2004). Physical and psychological partner aggression across a decade: A growth curve analysis. Violence and Victims, 19, 3-16.

18.

Furman

Brown

B. B.

Feiring

(1999). The development of romantic relationships in adolescence. New York, NY: Cambridge University Press.

19.

Furstenberg

(2001). The fading dream: Prospects for marriage in the inner city. In Anderson

Massey

D. S.

(Eds.), Problem of the century: Racial stratification in the United States (pp. 224-246). New York, NY: Russell Sage Foundation.

20.

Furstenberg

Morgan

S. P.

Moore

K. A.

Peterson

J. L.

(1987). Race differences in the timing of adolescent intercourse. American Sociological Review, 52, 511-518.

21.

Giordano

P. C.

Copp

J. E.

Longmore

M. A.

Manning

W. D.

(2015). Contested domains, verbal “amplifiers,” and intimate partner violence in young adulthood. Social Forces. Advance online publication. doi:10.1093/sf/sov055

22.

Giordano

P. C.

Soto

D. A.

Manning

W. D.

Longmore

M. A.

(2010). The characteristics of romantic relationships associated with teen dating violence. Social Science Research, 39, 863-874.

23.

Harknett

Kuperberg

(2012). Education, labor markets and the retreat from marriage. Social Forces, 90, 41-64.

24.

Hettrich

E. L.

O’Leary

K. D.

(2007). Females’ reasons for their physical aggression in dating relationships. Journal of Interpersonal Violence, 22, 1131-1143.

25.

Holtzworth-Munroe

Stuart

G. L.

Hutchinson

(1997). Violent versus nonviolent husbands: Differences in attachment patterns, dependency, and jealousy. Journal of Family Psychology, 11, 314-331.

26.

Johnson

W. L.

Giordano

P. C.

Manning

W. D.

Longmore

M. A.

(2015). The age-IPV curve: Changes in the perpetration of intimate partner violence during adolescence and young adulthood. Journal of Youth and Adolescence, 44, 708-726.

27.

Kerr

D. C. R.

Capaldi

D. M.

(2011). Young men’s intimate partner violence and relationship functioning: Long-term outcomes associated with suicide attempt and aggression in adolescence. Psychological Medicine, 40, 759-769.

28.

Kim

H. K.

Laurent

H. K.

Capaldi

D. M.

Feingold

(2008). Men’s aggression toward women: A 10-year panel study. Journal of Marriage and Family, 70, 1169-1187.

29.

Leone

J. M.

Johnson

M. P.

Cohan

C. L.

Lloyd

S. E.

(2004). Consequences of male partner violence for low-income minority women. Journal of Marriage and Family, 66, 472-490.

30.

Lichter

D. T.

McLaughlin

D. K.

Ribar

D. C.

(2002). Economic restructuring and the retreat from marriage. Social Science Research, 31, 230-256.

31.

Lichter

D. T.

Qian

(2008). Serial cohabitation and the marital life course. Journal of Marriage and Family, 70, 861-878.

32.

Manning

W. D.

Smock

P. J.

(2005). Measuring and modeling cohabitation: New perspectives from qualitative data. Journal of Marriage and Family, 67, 989-1002.

33.

Manning

W. D.

Trella

Lyons

Du Toit

N. C.

(2010). Marriageable women: A focus on participants in a community healthy marriage program. Family Relations, 59, 87-102.

34.

Meier

Allen

(2009). Romantic relationships from adolescence to young adulthood: Evidence from the National Longitudinal Study of Adolescent Health. Sociological Quarterly, 50, 308-335.

35.

Miller

White

N. A.

(2003). Gender and adolescent relationship violence: A contextual examination. Criminology, 41, 1207-1248.

36.

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, 942-949.

37.

Nomaguchi

K. M.

Giordano

P. C.

Manning

W. D.

Longmore

(2011). Adolescents’ gender mistrust: Variations and implications for the quality of romantic relationships. Journal of Marriage and Family, 73, 1032-1047.

38.

Patterson

(1998). Rituals of blood: Consequences of slavery in two American centuries. New York, NY: Basic Civitas.

39.

Payne

K. K.

(2014). The marriage rate and the Great Recession (FP-14-18). Retrieved from https://www.bgsu.edu/content/dam/BGSU/college-of-arts-and-sciences/NCFMR/documents/FP/FP-14-18-marriage-rate-recession.pdf

40.

Rauer

A. J.

Pettit

G. S.

Lansford

J. E.

Bates

J. E.

Dodge

K. A.

(2013). Romantic relationship patterns in young adulthood and their developmental antecedents. Developmental Psychology, 49, 2159-2171.

41.

Ross

C. E.

Mirowsky

Pribesh

(2001). Powerlessness and the amplification of threat: Neighborhood disadvantage, disorder, and mistrust. American Sociological Review, 66, 568-591.

42.

Ross

J. M.

(2011). Personality and situational correlates of self-reported reasons for intimate partner violence among women versus men referred for batterers’ intervention. Behavioral Sciences & the Law, 29, 711-727.

43.

Schaller

(2012). For richer, if not for poorer? Marriage and divorce over the business cycle. Journal of Population Economics, 26, 1007-1033.

44.

Smith

C. A.

Ireland

T. O.

Park

Elwyn

Thornberry

T. P.

(2011). Intergenerational continuities and discontinuities in intimate partner violence: A two-generational prospective study. Journal of Interpersonal Violence, 26, 3720-3752.

45.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The revised Conflict Tactics Scales (CTS2). Journal of Family Issues, 17, 283-316.

46.

Treas

Giesen

(2000). Sexual infidelity among married and cohabiting Americans. Journal of Marriage and the Family, 62, 48-60.

47.

Umberson

(1992). Gender, marital status, and the social control of health behavior. Social Science & Medicine, 34, 907-917.

48.

Vest

J. R.

Catlin

T. K.

Chen

J. J.

Brownson

R. C.

(2002). Multistate analysis of factors associated with intimate partner violence. American Journal of Preventive Medicine, 22, 156-164.

49.

Wilkinson

D. L.

Hamerschlag

S. J.

(2005). Situational determinants in intimate partner violence. Aggression and Violent Behavior, 10, 333-361.

50.

Wilson

W. J.

(1987). The truly disadvantaged: The inner city, the underclass, and public policy. Chicago, IL: University of Chicago Press.

51.

Wilson

W. J.

(1996). When work disappears. Political Science Quarterly, 111, 567-595.