Intimate Partner Violence (IPV) is a highly prevalent public health challenge, tied to lasting negative consequences for physical and psychological health, parenting and child development, and economic stability. Typically conceptualized as a private form of violence, most efforts to predict IPV have focused on individual-level risk markers (e.g., substance use, childhood experiences) or couple-level risk markers (e.g., relationship satisfaction, communication skills). However, because individuals and couples are affected by the environments they inhabit, the contexts within which couples operate may also impact the likelihood that a couple will experience or engage in IPV. Through three studies, this dissertation aimed to examine contextual predictors of IPV, as well as the ways in which such contextual predictors can exacerbate or decrease the risk of individual and dyadic predictors on IPV. In an effort to synthesize prior work, Study 1 examined whether the accumulation of selected factors across individual, relational, and contextual socio-ecological layers, when considered simultaneously, predicts IPV. Results showed that even after adjusting for macro-level contextual influences (e.g., neighborhood and social network factors), individual and dyadic variables presented clear risk factors of IPV initial status. Associations between contextual variables and IPV were less robust, hinting at the possibility that the macro-contexts assessed in Study 1 may be less predictive of IPV than micro-level contextual factors. Therefore, the goal of Study 2 was to examine the effects of three micro-level contexts – perceived stress, financial strain, and experiences of discrimination. Specifically, Study 2 tested whether adversities experienced early in life serve to channel individuals into stressful circumstances (i.e., micro-contexts) that then evoke situational IPV in adulthood. Among husbands, early adversity was linked to IPV via stress, whereas for wives, no such mediation emerged. These findings indicate that the situations that are a defining feature of situational IPV may themselves be a reflection of the adversities that men face early in life; in the absence of these stressors, the association between early adversity and later IPV falls to non-significance. Finally, Study 3 of this dissertation examined whether the well-established association between psychological and physical IPV is moderated by the demands imposed upon couples by living in socially and economically disadvantaged contexts. Findings indicate that psychological and physical IPV were more likely to co-vary among husbands facing higher levels of socioeconomic disadvantage. I also tested whether negative and ineffective communication during relationship-focused conversations would moderate the association between psychological and physical IPV. Behavioral processes did indeed moderate this association, and the effect of behavioral processes was independent of the moderating effect of sociodemographic risk. Across the three studies, findings challenge the notion that IPV should be conceptualized exclusively as a private phenomenon. Instead, results lend general support to the value of understanding couples within their larger ecological niches and underscore the idea that that contextual risk factors, in addition to individual and relational variables, have the potential to influence whether couples’ arguments take on aggressive or even violent forms.

Although the consequences of divorce are felt across the economic spectrum, these costs are particularly high among low-income couples who experience relationship dissatisfaction and divorce at a disproportionate rate. Unfortunately, intervention efforts to improve marriages among low-income couples have proven unsuccessful, likely because these interventions assumed that well-established predictors of relationship distress studied primarily among middle-class samples would generalize to low-income couples. The three studies in this dissertation draw upon well-established predictors of relationship distress—poor communication skills and the propensity to blame a partner—in an effort to evaluate how those concepts may operate differently after accounting for the stressors and resources within low-income couples’ environment. Studies 1 and 2 examine whether couples’ behaviors under certain conditions carried greater consequences for their relationship satisfaction than under other conditions. Results of Study 1 indicate that when couples’ environments changed and they moved to poorer or more affluent neighborhoods, the impact of the change on relationship satisfaction was dependent on couples’ communication behaviors. Although couples with varying levels of communication skills did not differ from one another when they moved to similar neighborhoods, constructive communication behaviors did predict which couples thrived versus failed when faced with the stress of moving to a more affluent neighborhood environment. Study 2 extends the first study by examining fluctuations in behavior within a given couple and by examining other contextual factors that may impinge on their ability to cope: stress from work, discrimination, and finances. Results of Study 2 indicate that downward fluctuations in effective problem-solving behavior predicted downward fluctuations in satisfaction only among wives undergoing high levels of chronic stress, whereas decreases in effectiveness were inconsequential among wives experiencing little stress. Study 3 addresses how another classic predictor of relationship satisfaction—couples’ maladaptive attributions—may function differently when couples face high cognitive demands that may make it more difficult to consciously refrain from blaming attributions. Specifically, Study 3 tests whether strain from finances and potential compensatory resources from extended social networks (i.e., financial and social capital and models of intact marriages) are associated with couples’ maladaptive attributions and whether these factors serve to exacerbate or attenuate the negative effects of maladaptive attributions on couples’ relationship satisfaction. Results of Study 3 indicate that spouses who reported having greater financial strain and who knew fewer married individuals were more likely to form maladaptive attributions for their marital issues and that the affluence and marital status of family and friends attenuated the negative effects of maladaptive attributions on spouses’ relationship satisfaction. Thus, although communication behaviors and cognitive appraisals may very well be a fundamental process in all intimate relationships, the results of this dissertation cast doubt on the assumption that the behaviors and inferences couples make affect their relationship uniformly or universally.

Communication plays a key role in promoting the success of intimate relationships, but counterintuitive findings suggest that basic questions remain about how communication successfully promotes healthy relationships. This gap in understanding is likely due to three chief problems with current approaches to studying communication, all of which may inadequately describe the phenomenon: 1) Behaviors are traditionally characterized as either positive or negative, 2) Behavior is typically studied at the individual level, instead of as dyadic interactions, 3) Behavior is typically analyzed at the global, summative level, without acknowledging the most important momentary events within the interaction. The current dissertation proposed a new framework and accompanying coding system for studying communication that relies on two dimensions of behavior: cooperation (i.e., task-focused behavior) and affiliation (i.e., relationship-focused behavior). This work proposed that affiliation is the behavioral dimension that dictates the success of communication in promoting relationship quality, while cooperation is only associated with relationship wellbeing to the extent that it is accompanied by affiliation or disaffiliation. Study 1 tested these ideas at the individual mean level, while also examining how dyadic sequences of interaction predict relationship satisfaction over time. At the individual level, affiliation and its interaction with cooperation were associated with relationship satisfaction cross-sectionally, while dyadic sequences of sustained affiliative or disaffiliative reciprocity distinguished between levels of satisfaction cross-sectionally and accounted for variability in trajectories over time. Study 1 suggests that relationship-focused behavior is the key element of communication that accounts for variability in relationship satisfaction levels and trajectories. Study 2 of this dissertation challenges the notions that all behaviors occurring within an interaction contribute equally to relationship satisfaction, and instead draws from the peak-end rule (Kahneman, 2000) to propose that communication quality at three brief moments in couple conversations—namely, at its most positive peak, at its least negative valley, and at the ending moments of the conversation—may be more strongly associated with partners’ global judgments of relationship satisfaction. Higher peak values of husbands’ cooperative behavior were associated with slower declines in wives’ satisfaction, while higher valley values of husbands’ affiliative behavior slowed declines in husbands’ satisfaction, and higher ending values of wives’ affiliative behavior slowed declines in wives’ satisfaction. In sum, specific moments in couples’ conversations appear to have greater predictive value than the overall aggregated valence of their behavior. These two studies reveal that considering relationship-focused behaviors, dyad-level processes, and uniquely meaningful moments of interaction is essential to understanding the constructive or destructive nature of communication processes.

Although intimate relationships are a rewarding experience for many, others struggle to maintain healthy and lasting partnerships. This is especially true of couples living with low-incomes, who experience adverse relationship outcomes, such as divorce and dissatisfaction, at a disproportionate rate. Despite the need for effective interventions to prevent negative outcomes for lower-income couples and their families, the existing literature is equivocal on the most effective ways to intervene. Using data taken from three randomized controlled trials, the three studies in this dissertation project serve to test existing models of intervention, focusing on enhancing our knowledge of how these interventions operate and for whom they work best. Study 1 examines whether intervention effects vary systematically as a function of risk in three theoretically distinct preventive programs. Results indicate that treatment effects varied as a function of risk, and more so with variables capturing relational risk factors than individual risk factors. High-risk couples (e.g., couples with lower levels of baseline commitment and satisfaction) tended to decline less rapidly in satisfaction than low-risk couples following treatment. Study 2 examines whether intervention effects on relationship satisfaction are mediated by observational assessments of relationship communication and whether any such effects are moderated by couples’ pretreatment risk. Results indicate that couples who received the intervention reported higher average satisfaction at 30 months than control couples, regardless of their level of pretreatment risk, but only higher risk couples experienced an improvement in observed communication. Contrary to prediction, treatment effects on satisfaction were not mediated by improvements in communication. Finally, Study 3 examines the effects of job-related and school-related interventions on 3-year marriage rates, and whether relational or financial factors mediate this effect. Results indicate that marriage rates decreased, from 17% to 10%, among couples in which men participated in school-related interventions. Mediation analyses indicate that school-related interventions reduce the amount of time men spend with their child and the amount of money they contribute to their household, reducing marriage rates in turn. Together, these studies help refine our understanding of how to best improve relationship outcomes through preventive programs.

This project employs a developmental framework guided by interpersonal theories of depression and the transmission of intimate relationship dysfunction to offspring. We use two datasets to consider the independent and interactive impact of depression history, parental divorce and conflict, and relationship quality with parents on subsequent intimate relationship functioning from adolescence throughout the first eight years of marriage. The first study, using a longitudinal sample of adolescents, determined that participant/maternal relationship quality and parental marital stability predicted relationship conflict and satisfaction approximately 13 years later. Furthermore, adolescent depression history amplified the effect of some family-of-origin variables on some relationship outcomes. Study 2, using a longitudinal sample of newlywed couples, found that neither depression history nor family-of-origin variables predicted or interacted to enhance risky partner selection (with partner's risk defined through a factor- analyzed construct of emotion dysregulation). Instead, one's own level of risk was a strong predictor of partner's risk, supportive assortative mating theories. Study 3, again using the longitudinal newlywed sample, examined which components of intimate relationship communication predicted depressive symptoms approximately eight years later. Results demonstrated that participant's post-interaction evaluations of a negative mood were more consistent predictors of future depression than externally-rated communication behaviors, affect or skills, for both men and women. In addition, among men, a history of clinical or subclinical depression amplified the association between participant's negative evaluations of marital interactions and subsequent depressive symptoms. Taken together, these results suggest that factors well before relationship entry (i.e., psychopathology, familial functioning, emotion dysregulation) impact the quality of later intimate relationships, as well as characteristics of selected relationship partners. Furthermore, we provide evidence supporting attending to participant's evaluations and interpretations of marital interactions as predictors of later depressive symptoms beyond the content of this interpersonal communication.

Objective

Method

Results

Conclusions