Relationship-Specific Hope and Constructive Conflict Management in Adult Romantic Relationships: Testing an Accommodation Framework

Abstract

Using sequenced conflict interaction scenarios, this study tested Rusbult and colleagues’ partner accommodation framework. In addition, we examined the unique effects of relationship-specific hope, a variable we argue contributes to constructive conflict communication in adult romantic relationships. Results generally, but not completely, supported the predictive power of the partner accommodation model. Consistent with predictions, relationship-specific hope positively predicted partner accommodation in multiple scenarios. Moreover, relationship-specific hope positively predicted postconflict relational evaluations through its effects on partner accommodation. Supporting the tenets of hope theory, relationship-specific hope was found to be a more robust predictor of accommodation in multiple scenarios when partner responses were nonconciliatory (compared with conciliatory). This suggests that hope is most consequential to relational communication in especially trying relational circumstances. We interpret our findings, and their implications, based on interdependence theory, hope theory, and existing conflict management research.

Keywords

accommodation commitment conflict management hope investment model

How romantic partners manage conflicts can be a powerful indicator of a relationship’s health, functioning, and long-term viability. Many models of conflict management in personal relationships exist to explain the factors that promote constructive conflict communication (see Canary, 2003; Caughlin & Vangelisti, 2006; Roloff & Chiles, 2011). Rusbult, Verette, Whitney, Slovik, and Lipkus (1991) offered an especially heuristic model of “partner accommodation” that clarified how and why partners respond constructively to one another’s destructive actions. The accommodation framework, which was developed based on interdependence theory (Thibaut & Kelley, 1959) and the investment model (Rusbult, 1980), has guided much research because the framework identifies key antecedents (e.g., commitment) and consequences (e.g., partner functioning) of constructive conflict behaviors. This study further examines the accommodation model in adult romantic relationships, while also testing the effects of an additional variable that we believe is important in relational conflict management: relationship-specific hope.

Hope is the belief that the future holds promise and that goals are attainable, even if obstacles arise to block one’s goal pursuit (Snyder, 2002). Relationship-specific hope, by extension, reflects how romantic partners perceive and pursue the goals they have for their relational future. Partners with high relationship-specific hope can clearly identify relational goals. Moreover, they believe relational goals are achievable, despite obstacles, such as relational conflict (Merolla, 2014). Although hope is widely studied in health care (e.g., Farran, Herth, & Popovich, 1995) and education (e.g., Curry, Snyder, Cook, Ruby, & Rehm, 1997), and is gaining increased attention from communication scholars (Barge, 2003; Chadwick, 2015; Davis, 2013; Nabi & Keblusek, 2014), hope seems underutilized in the study of relational communication processes (Miller, Roloff, & Reznik, 2014).

The current study had four main objectives. First, we tested the accommodation model in adult romantic relationships using sequenced conflict interaction scenarios. Second, we examined if relationship-specific hope, when situated within an accommodation framework, positively predicts constructive conflict behavior. Third, we assessed if the association between relationship-specific hope and accommodation is moderated by partner communication (e.g., conciliatory responses). Fourth, we tested if the effect of relationship-specific hope on relational functioning is mediated by accommodation.

Partner Accommodation

One of the most important findings in the study of romantic relationships is that relationship health and longevity depend on partners’ ability to break chains of negative reciprocity in conflict situations through constructive communication (Caughlin & Vangelisti, 2006; Gottman, 2011; Rusbult, Yovetich, & Verette, 1996; Salazar, 2015). Yet, these chains can be quite strong, given that when partners engage in hurtful behavior—be it subtle criticism or overt hostility—self-interested and defensive responses are more readily activated than are pro-relational and constructive responses (Yovetich & Rusbult, 1994). Thus, a key task of relationship researchers has been to identify the variables underlying partners’ ability to accommodate negative behavior. Partner accommodation is defined as “an individual’s willingness, when the partner has enacted a potentially destructive behavior, to (a) inhibit impulses to react destructively in turn and (b) instead behave in a constructive manner” (Rusbult, Bissonnette, Arriaga, & Cox, 1998, p. 74). Research indicates that accommodation contributes to relational quality over time (Rusbult, Bissonnette et al., 1998). This is not to say that cooperative and forgiving behavior is inherently positive and functional in conflict situations (Canary, 2003; Luchies, Finkel, McNulty, & Kumashiro, 2010), but rather that when relationship maintenance goals are salient, accommodative responses during conflicts typically support those goals.

Rusbult and colleagues’ (Rusbult, Bissonnette et al. 1998; Rusbult et al., 1996) accommodation model (see Figure 1) is rooted in interdependence theory (Thibaut & Kelley, 1959), with three core interdependence theory constructs underlying the model: a given matrix, an effective matrix, and transformation of motivation. According to interdependence theory, when one partner enacts destructive behavior, the other partner is inclined to reciprocate that negative behavior and disinclined to respond with constructive behavior (Yovetich & Rusbult, 1994). This pattern of default self-interest is indicative of the given matrix (Rusbult, Bissonnette et al., 1998). But partners do not always behave according to the given matrix (Yovetich & Rusbult, 1994). Indeed, in ongoing relationships, partners often transform their motivation and engage in behavior indicative of the effective matrix. That is, they forgo self-interest and instead respond to partner negativity with constructive behavior.

Figure 1.

Depiction of the partner accommodation process.

To measure accommodation, Rusbult et al. (1991) developed the exit-voice-loyalty-neglect (EVLN) typology. This is a 2 × 2 matrix of behavior based on the dimensions of activeness-passiveness (i.e., degree to which partners directly or indirectly communicate their feelings about an offense) and constructiveness-destructiveness (i.e., the degree to which partners’ behavior following an offense supports relational well-being). Exit is an active and destructive response, voice is an active and constructive response, loyalty is a passive and constructive response, and neglect is a passive and destructive response. Accommodation, then, is operationalized as the use of voice and loyalty behaviors, and the disuse of exit and neglect behaviors, following partner misbehavior.

Rusbult, Bissonnette et al. (1998) proposed that the “macromotive” of accommodation in relationships is commitment level—a subjective estimate of partners’ dependence on one another and orientation toward relationship persistence. Furthermore, based on the interdependence theory (Thibaut & Kelley, 1959), as well as the investment model (Rusbult, 1980), Rusbult and colleagues (1991) proposed that commitment is built upon three primary relational factors: satisfaction (i.e., happiness with one’s relationship), investment size (i.e., time, resources, and emotional energy spent on a relationship), and quality of alternatives (i.e., opportunities available if one leaves the relationship). As shown in Figure 1, commitment is expected to be positively predicted by satisfaction and investment, and negatively predicted by quality of alternatives (Rusbult, Bissonnette et al., 1998). The model further proposes that the effects of commitment on partner functioning (i.e., evaluation of relationship health) are mediated by partners’ willingness to accommodate during conflict episodes.

Despite the accommodation model’s heurism and excellent predictive power (Agnew & Le, 2015), Rusbult and colleagues (Rusbult, Bissonnette et al. 1998; Rusbult et al., 1996) noted that the model is incomplete; many other proximal and distal predictors of accommodation exist beyond those in the model, including social norms, emotions, and individual dispositions (see Agnew & Le, 2015). Moreover, as Hui, Finkel, Fitzsimmons, Kumashiro, and Hofmann (2014) recently argued, commitment is often (but not always) a good predictor of accommodation. Many other factors must therefore be taken into account to more fully explain why and when accommodation transpires. To this end, Hui et al. (2014) suggested that the ways in which partners conceptualize and pursue long-term goals are critical for understanding accommodative behavior. For this reason, we believe that hope, which specifically accounts for goal-based cognition and behavior (Snyder, 2002), should be tested as a unique contributor to accommodation in romantic relationships. But what is hope and how has it been conceptualized in social scientific research? To answer this question, we briefly review social scientific perspectives on hope. We then outline our conceptualization of relationship-specific hope, and its consequences for relational communication.

Hope

Overview

Hope has garnered interest from writers in many disciplines and vocations for centuries. Early writing on hope appears in religious texts as well as ancient mythology, such as the tale of Pandora’s Box (Eliott, 2005; Snyder, 2000). Hope became a focal topic of social scientific research, however, much more recently—roughly in the middle part of the twentieth century (e.g., Menninger, 1959; Stotland, 1969). Hope is multifaceted (Farran et al., 1995; Webb, 2007), but many social scientists agree that hope is goal-directed. Thus, whereas everyday usage of the word “hope” often reflects a passive sense of waiting for good things to happen, social scientists often conceive hope in terms of the ways that people envision, pursue, and achieve their goals (Peterson & Byron, 2008). And of all the goal-based perspectives on hope, the most influential has been Snyder’s (2000, 2002) hope theory.

Hope theory

Hope theory defines hope as “a cognitive set that is based on a reciprocally-derived sense of successful agency (goal-directed determination) and pathways (planning to meet goals)” (Snyder et al., 1991, p. 571). Pathways are the routes people develop to pursue their goals, whereas agency is the motivation to pursue those routes. Hope, then, is equivalent to possessing well-developed pathways thinking and high motivational agency to pursue future goals (Snyder, 2002). Research shows that pathways and agency positively predict numerous outcomes, including academic achievement, business acumen, natural disaster recovery, disease management, and general well-being (e.g., Curry et al., 1997; Peterson & Byron, 2008; Snyder, 2002; Stanton, Danoff-Burg, & Huggins, 2002).

Importantly, scholars have carefully differentiated hope from seemingly similar variables, such as optimism and self-efficacy. Consider, for instance, optimism, a term that is often used interchangeably with hope in both research and everyday speech (Bruininks & Malle, 2005). Hope and optimism share a focus on outcome expectancies (see Snyder, 2002), but hope, unlike optimism, possesses a unique dual-structure with dedicated pathways and agency components (Snyder, 2002; Strauss, Niven, McClelland, & Cheung, 2014). As Shorey, Snyder, Rand, Hockemeyer, and Feldman (2002) put it, “An optimist may believe that things will turn out as he or she wants but not necessarily possess the pathways to pursue and acquire goals” (p. 325). Several psychometric studies (e.g., Bryant & Cvengros, 2004; Magaletta & Oliver, 1999) as well as a meta-analysis (Alarcon, Bowling, & Khazon, 2013) demonstrate that hope and optimism measures load onto unique latent factors. Similar results also have been found for self-efficacy, which maps onto the agency (but not the pathways) element of hope (Magaletta & Oliver, 1999).

Snyder (2002) proposed that hope is learned, and not genetic, with the social learning germane to hope beginning early in life. Over time, the cognitive and emotional aspects of hope take the form of a relatively stable disposition (Snyder, 2002). Individuals’ average levels of pathways and agency thinking across situations is referred to as trait-like hope. Hope also exists as (a) a state (i.e., hope level at a given moment) and (b) a context-specific perception (i.e., average hope level in a specific domain; see Lopez, Ciarlelli, Coffman, Stone, & Wyatt, 2000). While existing context-specific estimates of hope capture hopefulness across all of a person’s interpersonal relationships (Campbell & Kwon, 2001), our interest here is hope pertaining to a single relationship.

Relationship-specific hope

Grounded in hope theory, we define relationship-specific hope as the motivation to conjure relational goals along with the sense of agency to pursue those goals in an ongoing relationship, despite inevitable individual-, relational-, and system-level stressors that impede relational goal pursuit. Given that hope has been found to be relevant to large and small goals on a day-to-day basis (Ong, Edwards, & Bergeman, 2006; Snyder, 2002), relationship-specific hope should predict constructive communication in a variety of relational situations (Merolla, 2014). Nevertheless, because hope theory makes predictions about when hope is most consequential to behavior, we next consider how and when relationship-specific hope shapes partner interaction.

The Consequences of Relationship-Specific Hope for Partner Interaction

Hope theory suggests that hope is most consequential to behavior in difficult, novel, and challenging situations (Peterson & Byron, 2008; Snyder, 2002). Relationship-specific hope might operate similarly, such that it matters most in moments of relational difficulty, such as in conflicts. During conflicts, it is difficult to focus on relationship goals due to partners’ tendencies to communicate according to the given matrix (Rusbult et al., 1996). Relationship-specific hope can support accommodation in these moments because it is marked by the belief that problems can be overcome (Merolla, 2014). Consequently, relationship-specific hope can encourage integrative, and discourage disengaging, conflict responses (Miller et al., 2014; Yovetich & Rusbult, 1994).

More specifically, relationship-specific hope can support accommodation during conflict by furnishing partners with behavioral pathways (e.g., constructive acts; Sillars & Canary, 2013) and concomitant agency to enact those pathways (e.g., conflict efficacy; Makoul & Roloff, 1998) toward satisfactory relational outcomes. In support of this possibility, research shows that self-efficacy and self-control, which are moderate to strong correlates of hope (Magaletta & Oliver, 1999; Snyder et al., 1991), contribute to constructive conflict management in personal relationships (Afifi & Afifi, 2009; Finkel & Campbell, 2001). Self-control is particularly important for accommodation, given that it takes significant executive control to suppress immediate retaliatory goals after a partner’s hurtful behavior (Finkel & Campbell, 2001; Pronk & Karremans, 2014).

Yet, for self-control to predict accommodation during conflicts, individuals must be able to conjure up pro-relational goals, and be motivated to pursue those goals (Pronk & Karremans, 2014). As Pronk and Karremans (2014) put it, “executive control may only be related to sacrificial behavior when people indeed have the goal to protect and maintain their relationship” (p. 174). Relationship-specific hope, then, can promote accommodation in conflict situations by supporting partners’ sense of pro-relational goal-mindedness. These ideas might help explain why hope has been associated with constructive conflict management behavior in past research (Merolla, 2014; Miller et al., 2014).

Merolla (2014), for instance, reported that trait-like hope predicted accommodative behavior in regard to hypothetical and generalized romantic conflict situations. Merolla also reported that relationship-specific hope (above and beyond a series of covariates) positively predicted pro-relationship goals and negatively predicted hostility-domination goals in recent conflicts. Relationship-specific hope was also positively associated with the use of problem-solving conflict styles and negatively associated with the use of compliance, withdrawal, and third-party assistance. Miller et al. (2014) focused on how feelings of hopelessness with regard to the resolvability of a specific conflict affect conflict management in dating and parent-child relationships. Hopelessness was related to mutual hostility and disengaging conflict behaviors. Moreover, hopelessness (along with mutual hostility and self demand-withdrawal) mediated the relationship between desire for change and relational disengagement. In sum, it seems plausible that relationship-specific hope is a relational strength that supports self-efficacy, self-control, and pro-relational goal pursuit in moments of relational hostility, thus promoting accommodative conflict responses.

But is relationship-specific hope a unique relational strength? More specifically, does it predict accommodation above and beyond commitment, which is the most well-established motive of accommodation (Rusbult, Bissonnette et al., 1998)? Commitment and relationship-specific hope share some similarities, but they are unique constructs. Commitment and relationship-specific hope are similar insofar as they both involve considerations of a relationship’s future (Rusbult & Buunk, 1993). The constructs differ, however, in the extent to which they focus on relational goals. When partners perceive high commitment levels, it can be inferred that the maintenance of the relationship over time is a salient goal for the partners. As described earlier, however, goals are far more central to the conceptualization of relationship-specific hope (Merolla, 2014).

Relationship-specific hope’s specific focus on goals distinguishes it from commitment, given that relationships involve multiple goals beyond just relational maintenance (Roberts & Robins, 2000). Indeed, individuals can be committed to maintaining their relationship, but not especially hopeful about the prospects of achieving other important goals related to the relationship (e.g., sustaining passion, meeting financial goals, and co-parenting effectively). Because relationship-specific hope takes into account partners’ perceptions of multiple relational goals, it should uniquely contribute to the likelihood of accommodation during conflict episodes. Interdependence theory suggests that partners’ willingness to engage in pro-relational sacrifice in difficult relational situations is predicated on wide-ranging goal considerations (Van Lange, Agnew, Harinck, & Steemers, 1997). When individuals lack hope regarding relationship goals, they should be less likely to endure the costs associated with the transformation of motivation underlying accommodation (Finkel & Campbell, 2001; Pronk & Karremans, 2014; Rusbult & Buunk, 1993). As Miller et al. (2014) stated, “individuals who believe it is impossible to achieve their goals may be less motivated to try and instead engage in behaviors harmful to relationships” (p. 577).

Conceptualizing Accommodation in the Current Study

In this study, we developed conflict scenarios that capture the accommodative or nonaccommodative nature of individual communicative turns. Specifically, we assessed initial accommodation responses and successive accommodation responses. Initial accommodation responses are partners’ immediate EVLN responses to a partner transgression. Successive accommodation responses transpire later in the conflict after the transgressing partner reacts to the initial accommodation response. The distinction between initial and successive accommodation is useful because it enables us to test how different types of responses from the transgressing partner (e.g., conciliatory vs. nonconciliatory) moderate the relationship between participants’ initial and successive accommodative tendencies. It is possible, for instance, that a partner who is initially nonaccommodative becomes increasingly accommodative later in an interaction when the transgressing partner apologizes for his or her hurtful behavior (Bachman & Guerrero, 2006; Rusbult, Kumashiro, Finkel, & Wildschut, 2002). The opposite appears true, however, when transgressing partners enact nonconciliatory forms of communication, such as excuses, justifications, or refusals (Schönbach, 1990).

In this study, we made predictions about antecedents and consequences of initial and successive accommodation in the context of romantic conflicts based not only on the previously mentioned variables (i.e., commitment, relationship-specific hope) but also on the moderating effects of conciliatory versus nonconciliatory partner responses. We are particularly interested in successive accommodation responses, though, because they occur after three communicative turns have transpired (i.e., Partner A’s transgressive turn, Partner B’s initial accommodation response, and Partner A’s conciliatory or nonconciliatory response). We outline our specific sets of hypotheses next.

Hypotheses

We proposed three sets of predictions, all of which are modeled in Figure 2. The first set of predictions concerns the main tenets of the partner accommodation model (see Figure 1). The second set focuses on the interrelations between initial and successive accommodation and partner responses. The third set of predictions pertains to relationship-specific hope.

Figure 2.

Hypothesized model of partner accommodation including relationship-specific hope.

Basic Accommodation Model Predictions (H1-H3)

Consistent with the partner accommodation model, we hypothesized that commitment is positively predicted by satisfaction (H1a) and investment size (H1b), and negatively predicted by quality of alternatives (H1c). Also consistent with the partner accommodation model, we anticipated that commitment, functioning as the “central macromotive” of accommodation (Rusbult, Bissonnette et al., 1998), would positively predict both initial accommodation (H2a) and successive accommodation (H2b). Finally, and in line with the Rusbult, Bissonnette et al.’s’ (1998) accommodation framework, we assumed that the relationship between commitment and favorable postconflict relational evaluations would be mediated by successive accommodation (H3).

Accommodation and Partner Response Predictions (H4-H7)

We also hypothesized that initial accommodation would positively predict successive accommodation (H4). It indeed makes sense that partners’ motivation to accommodate in one turn should shape their motivation to accommodate in successive turns. Yet, given the interdependent nature of communication in conflict situations, we also needed to consider the role of partner responses (Rusbult et al., 2002; Schönbach, 1990). Conciliatory and apologetic responses, as noted earlier, can promote accommodation (Rusbult et al., 2002). Thus, we predicted that when partner responses are conciliatory, they should positively predict successive accommodation, but when they are nonconciliatory, they should negatively predict accommodation (H5). It is also plausible that partner response moderates the relationship between initial and successive accommodation (H6). For example, a partner may initially accommodate, but not successively accommodate after hearing a partner deny culpability for his or her behavior. Finally, we proposed that conciliatory partner responses positively predict postconflict relational evaluations; if the partner apologizes, the participant will rate the relationship more favorably than if the partner denies, excuses, or justifies his or her role in the conflict (H7).

Relationship-Specific Hope Predictions (H8-H10)

Consistent with our earlier discussion, we hypothesized that relationship-specific hope positively predicts initial accommodation (H8a) and successive accommodation (H8b). Mirroring the hypothesized mediated relationship between commitment and postconflict relational evaluations (H3), we also predicted that the link between relationship-specific hope and postconflict evaluations is mediated by successive accommodation (H9).

Furthermore, given our assertion that relationship-specific hope matters most in difficult situations, we hypothesized that relationship-specific hope would be a stronger predictor of successive accommodation when offending partners are nonconciliatory (relative to conciliatory; H10). Indeed, when transgressing partners are unremorseful, accommodation is especially difficult (Rusbult et al., 2002). Consistent with hope theory (Snyder, 2002), it is in this type of trying situation that relationship-specific hope should most strongly influence accommodation.

Method

Participants

This study involved three pilot studies to develop the sequenced conflict scenario methodology (described in more detail later) and one main data collection to test the hypotheses. Respondents for the main data collection were recruited from Mechanical Turk (MTurk). Participants in the main study had to be currently living in the United States and be in a romantic relationship. Participants received US$0.50 for completing the survey. The sample for this study was 367 adults (M_age = 30.64, SD_age = 9.33, range = 18-70 years, 45% male, 55% female). Approximately 55% of the sample identified as non-Hispanic White/Caucasian, 28% Asian/Pacific Islander, 7% Hispanic, 4% Native American, 4% African American/Black, and 3% Other. About half (53%) of participants were in dating relationships, 34% were married, 10% were engaged, 3% were in civil unions or domestic partnerships, and 1% reported being in an “other” relational form.

Materials and Measures

The data for the main study were collected through Likert-type measures as well as through responses to three conflict scenarios. Participants responded to each of the three scenarios developed for this study.

Investment model variables

The commitment, satisfaction, investment, and quality of alternatives measures came from Rusbult, Martz, and Agnew (1998), and used a 7-point response scale (1 = strongly disagree to 7 = strongly agree). The four items measuring commitment included statements, such as “I am committed to maintaining my relationship with my partner” (α = .89, M = 6.10, SD = 1.04). Satisfaction was measured with four items (e.g., “My relationship makes me very happy”; α = .92, M = 5.72, SD = 1.26). Four items also assessed investment size (e.g., “I have put a great deal into our relationship that I would lose if the relationship were to end”). Due to factor analysis (described below), two investment items were eventually dropped, making this a two-item measure (r = .62, M = 5.15, SD = 1.18). Four items assessed quality of alternatives (e.g., “The people other than my partner with whom I might become involved with are very appealing.”), but because one item referenced “dating” in our survey (even though married and engaged partners were in the sample), only three items were ultimately used (α = .80, M = 3.54, SD = 1.53).

Relationship-specific hope

Relationship-specific hope was measured with six items (e.g., “I have a lot of ways around any problems I might face in my relationship,” “I energetically pursue the goals I have for my relationship”) that were based on existing state and context-specific hope measures (see Lopez et al., 2000). The items were slightly modified for use in ongoing romantic relationships (Merolla, 2014). Response options ranged from definitely false (1) to definitely true (8). One item (i.e., “I see my relationship as being pretty successful”) was not used due to concerns that it overlapped with relational satisfaction and commitment items.

The survey also included Snyder et al.’s (1991) Adult Dispositional Hope Scale, a 12-item 8-point scale that includes four agency items (α = .86, M = 5.90, SD = 1.29), four pathways items (α = .80, M = 6.03, SD = 1.13), and four filler items. We therefore conducted tests to ensure that the relationship-specific hope measure was unique from dispositional hope. A three-factor confirmatory factor analysis (CFA) that included the two dispositional hope factors (agency and pathways) and a single relationship-specific hope factor demonstrated moderate fit, χ²(62) = 205.86, p < .001, confirmatory fit index (CFI) = .94, incremental fit index (IFI) = .94, root mean square error of approximation (RMSEA) = .08 (90% confidence interval [CI] = [.07, .09]). One of the relationship-specific hope items, however, had a low factor loading of .51 (all other loadings were above .70). Removal of this item produced a three-factor model with good fit, based on commonly accepted standards (e.g., Hu & Bentler, 1999), χ²(51) = 152.65, p < .001, CFI = .96, IFI = .96, RMSEA = .07 (90% CI = [.06, .09]). These results supported the uniqueness of a four-item relationship-specific hope index (α = .86, M = 6.08, SD = 1.29).

Sequenced Conflict Scenario Development

As shown in Figure 3, the conflict scenario procedure involved five steps: (a) participants first read a specific scenario involving partner transgressions (these are described momentarily); (b) participants then reported what their initial response would be (based on four options) if that situation happened in their relationship; (c) participants next received a randomized partner response; (d) participants then reported what their successive accommodation response would be based on Likert-type items; and finally (e) participants reflected on the entire conflict episode and evaluated the effects it would have on their relationship.

Figure 3.

Conflict scenario methodology flowchart.

The conflict scenarios were developed across three separate pilot studies. Participants in the pilot studies were undergraduate students enrolled in one of two classes (social psychology and communication studies) at a Western University during the spring and fall of 2011. Average class size across the pilot studies was approximately 100, and extra credit was awarded. No demographic information was solicited to protect students’ identities.

Pilot Study 1: Transgression examples

To create realistic conflicts, we asked respondents (via a Qualtrics survey) to give one to three examples of transgressions consistent with 12 of Metts’s (1994) relational violation types (e.g., violations of confidence, infidelity, breaking promises). Participants generated hundreds of examples for each violation. Ultimately, we chose the following violations for our three scenarios, because they captured different types of conflicts and seemed like sufficiently common incidents in intimate relationships:

Violation Scenario 1 (“Bad Day”): Imagine you are having a bad day—things didn’t go well at school or work. You’re hoping you can talk about your day with your partner. Your partner, though, says he or she doesn’t want to hear about it, and isn’t acting very supportive.

Violation Scenario 2 (“Flirting”); Imagine you catch your partner flirting with another person at a party. While you do not say or do anything at the party, you are alone with your partner on your drive home.

Violation Scenario 3 (“Critical Comment”): Imagine that you and your partner were talking over dinner at a restaurant, and he or she made a negative comment about your body type and your physical attractiveness.

Pilot Study 2: Initial accommodation responses

We created initial EVLN response options for each scenario by asking participants (via Qualtrics): What would you say or do if the scenario happened? We received hundreds of responses, which we categorized based on the EVLN typology (Rusbult et al., 1991). We then selected examples that best matched the EVLN categories. The wording of the generated examples was altered so that grammar was consistent. Table 1 show the initial accommodation response options used in the main study.

Table 1.

Initial Accommodation Responses and Number of Participants (From the Main Study) Selecting Each Tactic.

Scenario	Voice (accommodative)	Loyalty (accommodative)	Exit (nonaccommodative)	Neglect (nonaccommodative)
1. Bad day	I’d reassert that I need to talk about this and could really use his or her support.	I’d change the subject and talk about something else with the hope that he or she will come around.	I’d say something like, “If you’re not here for me now, why should I ever be there for you?”	I’d make it clear through my body language that I’m disappointed in him or her, and change the subject to something completely unrelated so it’s obvious that I’m upset.
1. Bad day	n = 139 (38%)	n = 81 (22%)	n = 35 (10%)	n = 111 (30%)
2. Flirting	I’d ask my partner who he or she was talking to at the party, and discuss my concerns about his or her flirting behavior.	I’d talk about how nice the people at the party were, and that it looked like my partner was having a good time.	I’d say something like, “You were flirting at the party” and clearly display my anger.	I’d talk about what a bad time I had at the party and mention that I did not like any of the people there.
2. Flirting	n = 179 (49%)	n = 100 (27%)	n = 66 (18%)	n = 21 (6%)
3. Critical comment	I’d tell her or him to be kinder and more sensitive about the way he or she says such things.	I’d laugh it off and make a joke about it.	I’d tell him or her that there are many other people who find me attractive.	I’d eat less, get quiet, and make it obvious that I was mad.
3. Critical comment	n = 86 (24%)	n = 103 (28%)	n = 86 (24%)	n = 89 (25%)

Note. The sum of the ns for each scenario is different due to a few missing cases. Specifically, there was one missing case in Bad Day and Flirting scenarios, and three missing cases in the Critical Comment scenario.

Pilot Study 3: Partner account examples

To create realistic conciliatory and nonconciliatory partner responses, participants in the third pilot study were asked during a class session to write out as many examples as they could for what the transgressing partners might say in the three relational violations generated in Pilot Study 1. We compiled all of the responses and sorted them into one of Schönbach’s (1990) four account categories: concession, justification, refusal, or excuse. The partner responses we ultimately used in the scenarios (see Table 2) were chosen based on realism and likelihood of use.

Table 2.

Randomized Partner Responses in Each Scenario (and Percentage of Participants Receiving Each Response).

Scenario	Concession (conciliatory)	Refusal (nonconciliatory)	Excuse (nonconciliatory)	Justification (nonconciliatory)
1. Bad day	I’m so sorry I was not sensitive to you. I will try to be more sensitive in the future. (25%)	What’s your problem? I don’t come to you with every single problem that I have. (26%)	I had a bad day too. So, that’s why earlier today I was really distracted and didn’t notice that you were upset. (23%)	Look, I just don’t see the value of focusing on negative experiences and complaining about having a bad day. It’s not good for you or our relationship. (25%)
2. Flirting	I am so sorry if it looked like I was flirting with that person at the party. That was not what I intended, and I should have introduced you as my partner. (29%)	By the way you’re acting, it feels like you think I did something wrong at the party. I didn’t do anything wrong. (26%)	That person I was talking to at the party knows that I am with you, so there’s nothing to worry about. (25%)	I used to work with that person I was talking to at the party, and we have always had a good relationship. (20%)
3. Critical comment	You’re gorgeous to me, I am sorry I said it that way. (23%)	You must have heard me say something else. I didn’t make any negative comments about your attractiveness. (25%)	Maybe I should not have said it that way. I am just really tired right now and am a little crabby. (26%)	Well, you do look different than when we first met. I am only trying to let you know that you have changed and could work harder to get back to how you used to look. (25%)

Measurement in Main Study Scenarios Based on Pilot Studies

Initial accommodation measurement

Participants were presented with the four response options developed in the pilot study (see Table 1), and were asked to select the one response that they would most likely use. This approach is consistent with previous research (e.g., Yovetich & Rusbult, 1994), and had two advantages. First, it enabled participants to move through the scenarios quickly. Indeed, it would have taken considerably longer if we used Likert-type measures of initial and successive accommodation (see Figure 3). Second, it allowed for unique measurement of initial and successive accommodation. Employing identical measures in such close proximity could reduce response variability and diminish participants’ interest in the scenarios. For the model testing, initial accommodation was dummy coded. Selection of voice or loyalty was coded as “1” and selection of exit or neglect was coded as “0.” Although dichotomizing constructive and destructive responses oversimplifies conflict communication, it is aligned with extant theory and research on accommodation (e.g., Perunovic & Holmes, 2008; Rusbult et al., 1996; Yovetich & Rusbult, 1994).

Partner response measurement

The random partner responses (see Table 2) were dichotomized for the model tests. Concessions were coded as “1” to represent conciliatory responses, and excuses, justifications, and refusals were coded as “0” to represent nonconciliatory responses.

Successive accommodation measurement

Successive accommodation was measured with a modified 12-item version of Rusbult et al.’s (1991) accommodation scale (1 = very unlikely to 5 = very likely). To estimate total accommodation, the exit and neglect items were reverse scored and averaged with the voice and loyalty items, such that higher scores indicated greater willingness to accommodate (Scenario 1: α = 78, M = 3.54, SD = 0.56; Scenario 2: α = .78, M = 3.48, SD = 0.61; and Scenario 3: α = .78, M = 3.40, SD = 0.64).

Postconflict relational evaluation

The final step (i.e., Step 5 in Figure 3) involved participants reporting how the events in the scenario would change their perceptions of the relationship. Participants were presented with statements, such as, “My relationship with my partner would feel,” with response options ranging from significantly less secure (1) to significantly more secure (7). Other items assessed qualities such as partner trustworthiness and happiness with the relationship. Reliability for the nine-item measure was high across the scenarios (Scenario 1: α = .97, M = 4.06, SD = 1.18; Scenario 2: α = .98, M = 3.91, SD = 1.23; and Scenario 3: α = .98, M = 3.81, SD = 1.23).

Results

Data Preparation and Preliminary Analyses

Tests of the 10 hypotheses were conducted with AMOS 20, using maximum likelihood estimation. Parcels were created for the postconflict relational evaluation and successive accommodation latent factors items due to the large number of items comprising those measures; each latent variable was composed of three parcels. For the successive accommodation items, we employed domain representative parceling (Little, Rhemtulla, Gibson, & Schoemann, 2013). No variable had more than 3% missing cases; any missing data were treated with maximum likelihood estimation. Prior to testing the hypotheses, CFAs were conducted on the latent factors. Moderate fit was found in the initial CFAs. Two of the investment items, however, had low factor loadings (i.e., <.60). With those items removed, model fit for each scenario was satisfactory (Hu & Bentler, 1999)—Scenario 1: χ²(209) = 501.31, χ²/df = 2.40, p < .001, CFI = .95, IFI = .95, RMSEA = .06 (90% CI = [.06, .07]); Scenario 2: χ²(209) = 466.01, χ²/df = 2.23, p < .001, CFI = .96, IFI = .96, RMSEA = .06 (90% CI = [.05, .07]); Scenario 3: χ²(209) = 450.75, χ²/df = 2.16, p < .001, CFI = .96, IFI = .96, RMSEA = .06 (90% CI = [.05, .06]). Factor loadings were also sufficiently high (ranges = .68-.95, .66-.95, and .66-.97 for the Scenarios 1, 2, and 3 models, respectively). Note, the full correlation matrix is available upon request.

Model Testing

Model setup and analytic plan

Given that each scenario had separate scenario variables, we tested the hypothesized model in Figure 2 separately for each scenario. This approach provided three tests for the hypotheses. For each model, we controlled for participant age and gender given that people of different ages and genders might have unique relational histories and social expectations influencing their responses (Rusbult et al., 1991). We also controlled for past experience, as it could shape accommodation and/or postconflict relational evaluations. We measured past experience with a straightforward “yes” or “no” response to the question: “Has something like this ever happened in your current relationship?” Note, for each model, we controlled for the influence of age, gender, and past experience on all of the model variables, with the exception of partner response, as this variable was randomly presented to the participants and thus should not be correlated to the control variables. For the same reason, partner response was also modeled as uncorrelated to the other exogenous variables in the models. To test the moderation effects proposed in H6 and H10, two product terms were entered into the model (i.e., Initial Accommodation × Partner Response and Relationship-Specific Hope × Partner Response). Both product terms were correlated with the exogenous variables in the model as well as the covariates. The errors of the product terms were correlated because they both contained the partner response variable. Finally, the effect of initial accommodation on postconflict relational evaluations was controlled for in each model.

Model fit and hypothesized paths

As shown in Table 3, the model fit for each of the three scenarios was acceptable, though the fit was slightly better for the second and third scenarios than the first one. Table 4 provides all of the standardized path estimates for the hypothesized model in each scenario. H1 proposed that commitment would be positively predicted by (a) satisfaction and (b) investment and (c) negatively predicted by quality of alternatives. These predictions were fully supported (see Table 4). H2 stated that commitment would positively predict (a) initial and (b) successive accommodation. While commitment only marginally predicted initial accommodation in the Scenario 2 model (p < .10), it significantly predicted successive accommodation in Scenarios 1 and 3 (but not 2). H2 thus received partial support, primarily for successive accommodation.

Table 3.

Goodness of Fit Statistics for the Hypothesized Model in Each Scenario.

Model	df	χ²	χ²/df	IFI	CFI	RMSEA [90% CI]
Scenario 1 (bad day)	373	785.86	2.11	.95	.95	.06 [.050, .060]
Scenario 2 (flirting)	373	705.74	1.89	.96	.96	.05 [.044, .055]
Scenario 3 (critical comment)	373	715.69	1.92	.96	.96	.05 [.045, .056]

Note. All models were statistically significant at p < .001. IFI = incremental fix index; CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval.

Table 4.

Standardized Regression Weights for Hypothesized Model Paths in Each Scenario.

Model paths	Scenario 1 (bad day)	Scenario 2 (flirting)	Scenario 3 (critical comment)
Basic accommodation model predictions
Satisfaction → Commitment (H1a)	.455****	.455****	.457****
Investment → Commitment (H1b)	.428****	.429****	.425****
Alternatives → Commitment (H1c)	−.193****	−.189****	−.195****
Commitment → Int Accom (H2a)	.019	.114^†	.071
Commitment → Scv Accom (H2b and first path in H3)	.222***	.119^†	.184**
Scv Accom → Rel Eval (second paths of H3 and H9)	.206**	.325****	.283****
Initial and successive accommodation and partner response predictions
Int Accom → Scv Accom (H4)	.151**	.391****	.340****
Partner Resp → Scv Accom (H5)	.820**	.594*	.422^†
Int Accom × Partner Rep → Scv Accom (H6)	.025	.073	.031
Partner Resp → Rel Eval (H7)	.026	.145**	.081^†
Relationship-specific hope predictions
Rel Hope → Int Accom (H8a)	.151*	.001	.062
Rel Hope → Scv Accom (H8b and first path of H9)	.434****	.284****	.252***
Rel Hope × Partner Resp → Scv Accom (H10)	−.664*	−.655*	−.386
Control variables and nonhypothesized paths involving scenarios variables
Gender → Int Accom	−.030	.006	−.237****
Gender → Scv Accom	−.059	.055	−.110*
Gender → Rel Eval	−.085	−.128**	−.160****
Age → Int Accom	.090^†	.071	.023
Age → Scv Accom	.057	.104*	.076
Age → Rel Eval	−.096^†	−.098*	−.099*
Past Experience → Int Accom	.039	.077	−.138**
Past Experience → Scv Accom	.066	.033	−.050
Past Experience → Rel Eval	−.210****	−.260****	−.275****
Int Accom → Rel Eval	−.015	.043	.067
Commitment → Rel Eval	−.107	−.173**	−.130*
Rel Hope → Rel Eval	.179*	.136*	.235****

Note. Gender was coded as 1 = men and 2 = women. Past experience was coded as 1 = yes and 2 = no. Not all control variable paths are displayed to preserve space. Alternatives = quality of alternatives; Int Accom = initial accommodation; Scv Accom = successive accommodation; Rel Eval = postconflict relational evaluation; Partner Resp = partner response; Rel Hope = relationship-specific hope.

†

p < .10. *p < .05. **p < .01. ***p < .001. ****p < .0001.

H3 predicted that the positive link between commitment and postconflict relational evaluations is mediated by successive accommodation. To test for mediation, we used RMediation, which provides 95% confidence intervals for mediated effects (Tofighi & MacKinnon, 2011). The commitment → successive accommodation → relational evaluation path was significant in Scenario 1 (unstandardized indirect effect estimate = .055, SE = .03, 95% CI = [.01, .12]) and Scenario 3 (unstandardized indirect effect estimate = .061, SE = .03, 95% CI = [.02, .12]), but not Scenario 2. This partially supports H3. Note that for Scenario 3, there was evidence of “inconsistent mediation” (MacKinnon, Krull, & Lockwood, 2000), whereby commitment positively predicted postconflict relational evaluation indirectly through successive accommodation, but negatively predicted postconflict relational evaluations directly (see Table 4). This suggests that the critical comment scenario was viewed as more destructive in highly committed relationships (the negative direct effect). Yet, because commitment increases accommodation, and accommodation improves partner functioning (see Figure 1), commitment can also indirectly improve postconflict relational evaluations (the positive indirect effect).

H4 predicted that initial accommodation would positively predict successive accommodation; support for this was found in all three scenarios. H5 stated that conciliatory partner responses positively predict successive accommodation. This effect was found in Scenarios 1 and 2, but not Scenario 3 (p < .10). H6 received no support, as partner response did not significantly moderate the relationship between initial and successive accommodation in any of the scenarios. H7, which asserted that partner response would positively predict postconflict relational evaluations, was supported in Scenario 2 only.

H8 stated that relationship-specific hope would positively predict (a) initial and (b) successive accommodation. Relationship-specific hope significantly predicted initial accommodation in Scenario 1 only, but it significantly predicted successive accommodation in all three scenarios. This partially supports the first part of H8, and fully supports the second part. H9, which predicted that relationship-specific hope predicts postconflict relational outcomes through successive accommodation, was fully supported. Using RMediation (Tofighi & MacKinnon, 2011), we found significant mediation effects in Scenario 1 (unstandardized indirect effect estimate = .086, SE = .04, 95% CI = [.02, .16]), Scenario 2 (unstandardized indirect effect estimate = .089, SE = 0.03, 95% CI = [.04, .16]), and Scenario 3 (unstandardized indirect effect estimate = .066, SE = .03, 95% CI = [.02, .12]).

H10 predicted that the effect of relationship-specific hope on successive accommodation would be stronger when partners were nonconciliatory (vs. conciliatory). Significant moderation effects were found in Scenarios 1 and 2, thus partially supporting the hypothesis. Follow-up ordinary least squares (OLS) regressions showed that the standardized regression coefficients for relationship-specific hope predicting successive accommodation in Scenarios 1 and 2 were .30 (p < .001) and .19 (p < .01) when partner responses were nonconciliatory, but only .03 (ns) and .01 (ns) when partner responses were conciliatory.

Additional Analysis Involving the Covariates

There were a few notable effects involving the control variables of gender, age, and past experience. As shown in Table 4, women were less likely than men to accommodate (initially and successively) in the critical comment scenario. Women also reported less positive postconflict relational evaluations in the flirting and critical comment scenarios. Age consistently, if meagerly, predicted lower postconflict relational evaluations. Past experience also influenced postconflict relational evaluations such that in all three scenarios, participants who experienced similar transgressions in the past (compared with those who did not) reported more positive postconflict relational evaluations.

Discussion

In personal relationships, the amount of conflict that partners experience appears to be less consequential to their relational quality than the ways in which they manage that conflict (Sillars & Canary, 2013). Reducing escalating negativity is especially important (Gottman, 2011). Rusbult and colleagues (Rusbult, Bissonnette et al., 1998; Rusbult et al., 1991; Rusbult et al., 1996) provided one of the most heuristic, theory-based models explaining how partners can reduce negativity during conflicts through accommodative behavior. Accommodation is defined as the use of constructive responses (i.e., voice or loyalty) and the disuse of destructive responses (i.e., exit and neglect) following partner misbehavior. The current study utilized sequenced conflict interaction scenarios to further test the accommodation framework, and to extend it through the inclusion of relationship-specific hope.

Turning first to the basic accommodation model predictions, the pattern of results generally supported the model’s validity, with some notable exceptions. As expected, commitment was positively predicted by satisfaction and investment size, and negatively predicted by quality of alternatives. Commitment, in turn, was a significant positive predictor of successive accommodation. Also consistent with Rusbult, Bissonnette et al.’s (1998) theorizing, the relationship between commitment and postconflict relational evaluations was mediated by successive accommodation.

These direct and mediated effects of commitment, however, were only significant in Scenarios 1 and 3 (i.e., the “bad day” and “critical comment” scenarios). It is possible that commitment did not predict successive accommodation in Scenario 2—the flirting scenario—due to the nature of the offense. Factors not included in the model, such as attachment anxiety and self-concept, may better predict responses in betrayal situations like the flirting scenario (see Besser & Priel, 2009). Commitment also failed to perform as expected in regard to initial accommodation. Despite being positioned as the central motive of accommodation (Agnew & Le, 2015; Rusbult, Bissonnette et al., 1998), commitment did not significantly predict initial accommodation in any of the scenarios. This was unexpected, and, as we discuss later, could be due to the differing nature of initial and successive accommodation in conflict episodes or issues with our operationalization of initial accommodation.

Perhaps the most noteworthy findings from this study concern relationship-specific hope, which was found to be a significant and positive predictor of initial (Scenario 1) and successive (Scenarios 1, 2 and 3) accommodation. A significant mediation effect was also found in all three scenarios, whereby relationship-specific hope positively predicted postconflict relational evaluations through successive accommodation. Notably, the significant paths between relationship-specific hope and accommodation (initial and successive) were stronger than those between commitment and accommodation. Furthermore, and consistent with hope theory (Snyder, 2002), relationship-specific hope was a more robust predictor of successive accommodation (in Scenarios 1 and 2) when partner responses were nonconciliatory (compared with conciliatory). This finding is consistent with the perspective that hope matters most when obstacles block goals pursuits (Snyder, 2002). Indeed, when offending partners are nonconciliatory, the transformation of motivation underlying partner accommodation requires substantial effort (Rusbult et al., 2002).

Although the importance of hope has been acknowledged in the existing conflict literature at the relational (e.g., Fincham, Bradbury, & Grych, 1990) and societal (e.g., Jarymowicz & Bar-Tal, 2006) levels, hope is rarely measured in studies of relational conflict and communication (cf. Miller et al., 2014). We have argued that the partner accommodation model, with its strong emphasis on commitment as the central motive of constructive conflict behavior, does not fully account for whether or not partners perceive that they have the ability and motivation to reach the goals they have for the relationship (Finkel & Campbell, 2001; Hui et al., 2014). Our results provide evidence that relationship-specific hope could be a critical factor supporting partners’ willingness to accommodate, perhaps because a hopeful outlook enables partners to focus on relational-level rather than individual-level goals during difficult moments (Merolla, 2014).

Based on the tenets of interdependence theory, relationship-specific hope might support the transformation of motivation from given to effective matrix preferences (Rusbult et al., 1996) by providing partners with: (a) a variety of behavioral pathways for constructive conflict management and (b) requisite efficacy to enact those pathways during conflict episodes. These dual factors might help partners avoid intractable conflicts marked by chronic avoidance (Miller et al., 2014). In short, when partners are hopeful about their relationships, they are less likely to give up when difficult situations arise, and more likely to put in the effort necessary to transform their motivation (Braithwaite, Selby, & Fincham, 2011; Miller et al., 2014).

It indeed makes sense that relationship-specific hope supports the transformation of motivation, which is often an effortful and taxing experience that necessitates significant self-control (e.g., Finkel & Campbell, 2001; Hui, Molden, & Finkel, 2013; Pronk & Karremans, 2014). Unless partners perceive that they possess the pathways and agency necessary to foster desirable outcomes (Snyder, 2002) in the relationship, their inclinations toward accommodation will be reduced due to a perception that conflict resolvability is low (Johnson & Roloff, 2000; Malis & Roloff, 2006). Furthermore, low hope might increase negative emotional flooding during conflicts (Merolla, 2014; Miller et al., 2014), which can undermine the likelihood that partners can constructively respond to partner misbehavior (Gottman, 2011).

The current findings are also consistent with the literature on relational therapy. Existing hope-based relational enrichment programs, for example, have been shown to be effective at increasing partners’ relational satisfaction as well as their conflict communication skills (Worthington et al., 1997). Furthermore, in both relational and nonrelational contexts (e.g., academics), research demonstrates that long- and short-term hope theory-based interventions effectively increase individuals’ hope levels and positive outcomes (see, for example, Feldman & Dreher, 2012). Our results suggest that practitioners should continue to target hope in relational interventions focused on increasing constructive conflict management (Worthington et al., 1997).

Limitations and Future Research

We first want to revisit the results for initial accommodation responses. As noted earlier, although commitment and relationship-specific hope well-predicted successive accommodation, they were not good predictors of initial accommodation. We see two potential explanations for this. First, the effects could be attributable to the unique locations of initial and successive accommodation in the conflict. Because successive accommodation transpires further into the conflict (i.e., following the transgression, initial accommodation, and partner response), it is more “diagnostic” of partners’ pro-relational tendencies (Agnew & Le, 2015; Rusbult et al., 1996).

A second explanation concerns our operationalization of initial accommodation. Unfortunately, our measurement of initial accommodation involved tradeoffs. Two benefits of using the forced-choice items included that they (a) enabled quick responses from the participants, and (b) provided a way for us to uniquely measure initial and successive accommodation (which could maintain participants’ interest in the survey). Two drawbacks of our measurement, however, are that the participants (a) might not have identified with any of the four options presented to them or (b) interpreted the four options differently than we intended. For instance, a participant could have inferred a sarcastic tone in one of the response options (even if we did not intend for such a tone to be present). A related limitation regarding our measurement of initial accommodation comes from our decision to dichotomize constructive and destructive responses. Partner communication during conflict can be quite complex and serve multiple goals and functions (Caughlin & Vangelisti, 2006). Thus, our decision to collapse voice and loyalty response as “constructive” and exit and neglect responses as “destructive”—though consistent with Rusbult and colleagues’ (1996) conceptualization of accommodation—significantly reduced the representativeness of the wide-ranging communicative responses participants might enact in actual conflict episodes. Yet, despite the limited response options, results were often in line with expectations. In all three scenarios, for example, initial accommodation significantly and positively predicted successive accommodation (as would be expected). This alleviates some, but not all, of the concern over the validity of the initial accommodation measurement.

A major limitation of this study is our use of self-report and scenario data. Although we attempted to create realistic scenarios based on pilot-testing and novel features (e.g., randomized partner responses), we did not observe actual communication, and it is difficult to know how closely self-reported behavior matches actual behavior (cf. Rusbult et al., 1991). We also did not consider contextual factors that shape conflict (Caughlin & Vangelisti, 2006; Sillars & Canary, 2013). For example, if hope is most impactful on behavior in difficult circumstances, it could be most consequential to relational conflict during periods of high stress. Future research should explore these issues. Future research should also test whether (and how) relationship-specific hope benefits and detracts from relational functioning. Our finding that relationship-specific hope was a stronger predictor of successive accommodation when transgressing partners were nonconciliatory shows the positive consequences of hope for conflict management. On the other hand, it might also suggest that hopeful partners are vulnerable to overlooking serious relational inequity or dysfunction (Luchies et al., 2010). Studies tracking hope and relational behavior over time could be employed to test these ideas (see, for example, Ong et al., 2006).

Conclusion

According to hope theory (Snyder, 2002), it takes much more than positive expectations for people to attain their goals; people must also devise clear behavioral pathways toward their goals. The same is true in close relationships. In order to attain relational goals, such as maintaining relationship satisfaction over time, partners must manifest their positive expectations into constructive relational behavior (McNulty & Karney, 2004). Our results demonstrate some of the ways in which a hopeful relational outlook can manifest in constructive communication. Specifically, we found that relationship-specific hope promotes partner accommodation during conflict episodes. In this way, accommodative behaviors can be conceptualized as pathways through which partners can realize their broader relational goals. Integrating theory and research on hope (Snyder, 2002) and partner accommodation (Rusbult, Bissonnette et al., 1998), our findings suggest that relationship-specific hope might be a unique variable, at the intersection of cognition and communication, that contributes to well-functioning relationships over time.

Footnotes

Acknowledgements

The authors wish to thank Dr. Michael Roloff and the anonymous reviewers for their helpful guidance on this article.

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Andy J. Merolla (PhD, The Ohio State University) is an associate professor of Communication Studies in the Department of Communication Arts & Sciences at Baldwin Wallace University. His research examines how people maintain relationships in difficult circumstances.

Jennifer J. Harman (PhD, University of Connecticut) is an associate professor of Applied Social and Health Psychology in the Department of Psychology at Colorado State University. She specializes in the study of intimate relationships, particularly how intimate partners perceive and influence each other’s outcomes (e.g., mental and physical health).

References

Afifi

W. A.

Afifi

T. D.

(2009). Avoidance among adolescents in conversations about their parents’ relationship: Applying the theory of motivated information management. Journal of Social and Personal Relationships, 26, 488-511. doi:10.1177/0265407509350869

Agnew

C. R.

(2015). Prosocial behavior in close relationships: An interdependence approach. In Schroeder

D. A.

Graziano

W. G.

(Eds.), The Oxford handbook of prosocial behavior (pp. 363-375). New York, NY: Oxford University Press.

Alarcon

G. M.

Bowling

N. A.

Khazon

(2013). Great expectations: A meta-analytic examination of optimism and hope. Personality and Individual Differences, 54, 821-827. doi:10.1016/j.paid.2012.12.004

Bachman

G. F.

Guerrero

L. K.

(2006). Forgiveness, apology, and communicative responses to hurtful events. Communication Reports, 19, 45-56. doi:10.1080/08934210600586357

Barge

J. K.

(2003). Hope, communication, and community building. Southern Communication Journal, 69, 63-81. doi:10.1080/10417940309373279

Besser

Priel

(2009). Emotional responses to a romantic partner’s imaginary rejection: The roles of attachment anxiety, covert narcissism, and self-evaluation. Journal of Personality, 77, 1-39. doi:10.1111/j.1467-6494.2008.00546.x

Braithwaite

S. R.

Selby

E. A.

Fincham

F. D.

(2011). Forgiveness and relationship satisfaction: Mediating mechanisms. Journal of Family Psychology, 25, 551-559. doi:10.1037/a0024526

Bruininks

Malle

B. F.

(2005). Distinguishing hope from optimism and related affective states. Motivation and Emotion, 29, 327-355. doi:10.1007/s11031-006-9010-4

Bryant

F. B.

Cvengros

J. A.

(2004). Distinguishing hope and optimism: Two sides of a coin, or two separate coins? Journal of Social & Clinical Psychology, 23, 273-302. doi:10.1521/jscp.23.2.273.31018

10.

Campbell

D. G.

Kwon

(2001). Domain-specific hope and personal style: Toward an integrative understanding of dysphoria. Journal of Social and Clinical Psychology, 20, 498-520. doi:10.1521/jscp.20.4.498.22400

11.

Canary

D. J.

(2003). Managing interpersonal conflict: A model of events related to strategic choices. In Greene

J. O.

Burleson

B. R.

(Eds.), Handbook of communication and social interaction skills (pp. 515-549). Mahwah, NJ: Lawrence Erlbaum.

12.

Caughlin

J. P.

Vangelisti

A. L.

(2006). Conflict in dating and marital relationships. In Oetzel

J. G.

Ting-Toomey

(Eds.), The SAGE handbook of conflict communication: Integrating theory, research, and practice (pp. 129-157). Thousand Oaks, CA: SAGE.

13.

Chadwick

A. E.

(2015). Toward a theory of persuasive hope: Effects of cognitive appraisals, hope appeals, and hope in the context of climate change. Health Communication, 30, 598-611. doi:10.1080/10410236.2014.916777

14.

Curry

L. A.

Snyder

C. R.

Cook

D. L.

Ruby

B. C.

Rehm

(1997). Role of hope in academic and sport achievement. Journal of Personality and Social Psychology, 73, 1257-1267. doi:10.1037/0022-3514.73.6.1257

15.

Davis

C. S.

(2013). Communicating hope: An ethnography of a children’s mental health care team. Walnut Creek, CA: Left Coast Press.

16.

Eliott

(2005). What have we done with hope? A brief history. In Eliott

(Ed.), Interdisciplinary perspectives on hope (pp. 3-45). New York, NY: Nova.

17.

Farran

C. J.

Herth

K. A.

Popovich

J. M.

(1995). Hope and hopelessness: Critical clinical constructs. Thousand Oaks, CA: SAGE.

18.

Feldman

D. B.

Dreher

D. E.

(2012). Can hope be changed in 90 minutes? Testing the efficacy of a single-session goal-pursuit intervention for college students. Journal of Happiness Studies, 13, 745-759. doi:10.1007/s10902-011-9292-4

19.

Fincham

F. D.

Bradbury

T. N.

Grych

J. H.

(1990). Conflict in close relationships: The role of intrapersonal phenomena. In Graham

Folkes

V. S.

(Eds.), Attribution theory: Applications to achievement, mental health, and interpersonal conflict (pp. 161-184). Hillsdale, NJ: Lawrence Erlbaum.

20.

Finkel

E. J.

Campbell

K. W.

(2001). Self-control and accommodation in close relationships: An interdependence analysis. Journal of Personality and Social Psychology, 81, 263-277. doi:10.1037/0022-3514.81.2.263

21.

Gottman

J. M.

(2011). The science of trust: Emotional attunement for couples. New York, NY: W.W. Norton.

22.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1-55. doi:10.1080/10705519909540118

23.

Hui

C. M.

Finkel

E. J.

Fitzsimmons

G. M.

Kumashiro

Hofmann

(2014). The Manhattan effect: When relationship commitment fails to promote support for partners’ interests. Journal of Personality and Social Psychology, 106, 546-570. doi:10.1037/a0035493

24.

Hui

C. M.

Molden

D. C.

Finkel

E. J.

(2013). Loving freedom: Concerns with promotion or prevention and the role of autonomy in relationship well-being. Journal of Personality and Social Psychology, 105, 61-85. doi:10.1037/a0032503

25.

Jarymowicz

Bar-Tal

(2006). The dominance of fear over hope in the life of individuals and collectives. European Journal of Social Psychology, 36, 367-392. doi:10.1002/ejsp.302

26.

Johnson

K. L.

Roloff

M. E.

(2000). Correlates of the perceived resolvability and relational consequences of serial arguing in dating relationships: Argumentative features and the use of coping strategies. Journal of Social and Personal Relationships, 17, 676-686. doi:10.1177/0265407500174011

27.

Little

T. D.

Rhemtulla

Gibson

Schoemann

A. M.

(2013). Why the items versus parcels controversy needn’t be one. Psychological Methods, 18, 285-300. doi:10.1037/a0033266

28.

Lopez

S. J.

Ciarlelli

Coffman

Stone

Wyatt

(2000). Diagnosing for strengths: On measuring hope building blocks. In Snyder

C. R.

(Ed.), Handbook of hope: Theory, measures, & applications (pp. 57-85). San Diego, CA: Academic Press.

29.

Luchies

L. B.

Finkel

E. J.

McNulty

J. K.

Kumashiro

(2010). The doormat effect: When forgiving erodes self-respect and self-concept clarity. Journal of Personality and Social Psychology, 9, 734-749. doi:10.1037/a0017838

30.

MacKinnon

D. P.

Krull

J. L.

Lockwood

C. M.

(2000). Equivalence of the mediation, confounding and suppression effect. Prevention Science, 1, 173-181. doi:10.1023/A:1026595011371

31.

Magaletta

P. R.

Oliver

J. M.

(1999). The hope construct, will and ways: Their relations with self-efficacy, optimism, and general well-being. Journal of Clinical Psychology, 55, 539-551. doi:10.1002/(SICI)1097-4679(199905)

32.

Makoul

Roloff

M. E.

(1998). The role of efficacy and outcome expectations in the decision to withhold relational complains. Communication Research, 25, 5-29. doi:10.1177/009365098025001001

33.

Malis

R. S.

Roloff

M. E.

(2006). Demand/withdraw patterns in serial arguments: Implications for well-being. Human Communication Research, 32, 198-216. doi:10.1111/j.1468-2958.2006.00009.x

34.

McNulty

J. K.

Karney

B. R.

(2004). Positive expectations in the early years of marriage: Should couples expect the best or brace for the worst? Journal of Personality & Social Psychology, 86, 729-743. doi:10.1037/0022-3514.86.5.729

35.

Menninger

(1959). Hope. American Journal of Psychiatry, 116, 481-491. doi:10.1176/ajp.116.6.481

36.

Merolla

A. J.

(2014). The role of hope in conflict management and relational maintenance. Personal Relationships, 21, 365-386. doi:10.1111/pere.12037

37.

Metts

(1994). Relational transgressions. In Cupach

W. R.

Spitzberg

B. H.

(Eds.), The dark side of interpersonal communication (pp. 217-239). Hillsdale, NJ: Lawrence Erlbaum.

38.

Miller

C. W.

Roloff

M. E.

Reznik

R. M.

(2014). Hopelessness and interpersonal conflict: Antecedents and consequences of losing hope. Western Journal of Communication, 78, 563-585. doi:10.1080/10570314.2014.896026

39.

Nabi

R. L.

Keblusek

(2014). Inspired by hope, motivated by envy: Comparing the effects of discrete emotions in the process of social comparison to media figures. Media Psychology, 17, 208-234. doi:10.1080/15213269.2013.878663

40.

Ong

A. D.

Edwards

L. M.

Bergeman

C. S.

(2006). Hope as a source of resilience in later adulthood. Personality and Individual Differences, 41, 1263-1273. doi:10.1111/j.1467-6494.2009.00600.x

41.

Perunovic

Holmes

J. G.

(2008). Automatic accommodation: The role of personality. Personal Relationships, 15, 57-70. doi:10.1111/j.1475-6811.2007.00184.x

42.

Peterson

S. J.

Byron

(2008). Exploring the role of hope in job performance: Results from four studies. Journal of Organizational Behavior, 29, 785-803. doi:10.1002/job.492

43.

Pronk

T. M.

Karremans

J. C.

(2014). Does executive control relate to sacrificial behavior during conflicts of interests? Personal Relationships, 21, 168-175. doi:10.1111/pere.12024

44.

Roberts

B. W.

Robins

R. W.

(2000). Broad dispositions, broad aspirations: The intersection of personality traits and major life goals. Personality and Personal Psychology Bulletin, 26, 1284-1296. doi:10.1177/0146167200262009

45.

Roloff

M. J.

Chiles

B. W.

(2011). Interpersonal conflict: Recent trends. In Knapp

M. L.

Daly

J. A.

(Eds.), The SAGE handbook of interpersonal communication (4th ed., pp. 423-442). Los Angeles, CA: SAGE.

46.

Rusbult

C. E.

(1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16, 172-186. doi:10.1016/0022-1031(80)90007-4

47.

Rusbult

C. E.

Bissonnette

V. L.

Arriaga

X. B.

Cox

C. L.

(1998). Accommodation processes during the early years of marriage. In Bradbury

T. N.

(Ed.), The developmental course of marital dysfunction (pp. 74-113). Cambridge, UK: Cambridge University Press.

48.

Rusbult

C. E.

Buunk

B. P.

(1993). Interdependence in close relationships: An interdependence analysis. Journal of Social and Personal Relationships, 10, 175-204. doi:10.1177/026540759301000202

49.

Rusbult

C. E.

Kumashiro

Finkel

E. J.

Wildschut

(2002). The war of the roses: An interdependence analysis of betrayal and forgiveness. In Noller

Feeney

J. A.

(Eds.), Understanding marriage: Developments in the study of couple interaction (pp. 251-281). Cambridge, UK: Cambridge University Press.

50.

Rusbult

C. E.

Martz

J. M.

Agnew

C. R.

(1998). The investment model scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships, 5, 357-391.

51.

Rusbult

C. E.

Verette

Whitney

G. A.

Slovik

L. F.

Lipkus

(1991). Accommodation processes in close relationships: Theory and preliminary empirical evidence. Journal of Personality and Social Psychology, 60, 53-78. doi:10.1111/j.1475-6811.1998.tb00177.x

52.

Rusbult

C. E.

Yovetich

N. A.

Verette

(1996). An interdependence analysis of accommodation processes. In Fletcher

G. J. O.

Fitness

(Eds.), Knowledge structures in close relationships: A social psychological approach (pp. 63-90). Mahwah, NJ: Lawrence Erlbaum.

53.

Salazar

L. R.

(2015). The negative reciprocity process in marital relationships: A literature review. Aggression and Violent Behavior, 24, 113-119. doi:10.1016/j.avb.2015.05.008

54.

Schönbach

(1990). Account episodes: The management or escalation of conflict. Cambridge, UK: Cambridge University Press.

55.

Shorey

H. S.

Snyder

C. R.

Rand

K. L.

Hockemeyer

J. R.

Feldman

D. B.

(2002). Somewhere over the rainbow: Hope theory weathers its first decade. Psychological Inquiry, 13, 322-331. doi:10.1207/S15327965PLI1304_03

56.

Sillars

A. L.

Canary

D. J.

(2013). Conflict and relational quality in families. In Vangelisti

A. L.

(Ed.), The Routledge handbook of family communication (2nd ed., pp. 338-357). New York, NY: Routledge.

57.

Snyder

C. R.

(2000). Hypothesis: There is hope. In Snyder

C. R.

(Ed.), Handbook of hope: Theory, measures, and applications (pp. 3-21). San Diego, CA: Academic Press.

58.

Snyder

C. R.

(2002). Hope theory: Rainbows in the mind. Psychological Inquiry, 13, 249-275. doi:10.1207/S15327965PLI1304_01

59.

Snyder

C. R.

Harris

Anderson

J. R.

Holleran

S. A.

Irving

L. M.

Sigmon

S. T.

. . . Harney

(1991). The will and the ways: Development and validation of an individual-differences measure of hope. Journal of Personality and Social Psychology, 60, 570-585. doi:10.1037/0022-3514.60.4.570

60.

Stanton

A. L.

Danoff-Burg

Huggins

M. E.

(2002). The first year after breast cancer diagnosis: Hope and coping strategies as predictors of adjustment. Psycho-Oncology, 11, 93-102. doi:10.1002/pon.574

61.

Stotland

(1969). The psychology of hope: An integration of experimental, clinical, and social approaches. San Francisco, CA: Jossey-Bass.

62.

Strauss

Niven

McClelland

C. R.

Cheung

B. K. T.

(2014). Hope and optimism in the face of change: Contributions to task adaptivity. Journal of Business Psychology, 30, 733-745. doi:10.1007/s10869-014-9393-2

63.

Thibaut

J. W.

Kelley

H. H.

(1959). The social psychology of groups. New York, NY: John Wiley.

64.

Tofighi

MacKinnon

D. P.

(2011). RMediation: An R package for mediation analysis confidence intervals. Behavioral Research, 43, 692-700. doi:10.3758/s13428-011-0076-x

65.

Van Lange

P. A. M.

Agnew

C. R.

Harinck

Steemers

G. E. M.

(1997). From game theory to real life: How social value orientation affects willingness to sacrifice in ongoing close relationships. Journal of Personality and Social Psychology, 73, 1330-1344. doi:10.1037/0022-3514.73.6.1330

66.

Webb

(2007). Modes of hoping. History of the Human Sciences, 20, 65-83. doi:10.1177/095269510707933

67.

Worthington

E. L.

Hight

T. L.

Ripley

J. S.

Perrone

K. M.

Kurusu

T. A.

Hones

D. R.

(1997). Strategic hope-focused relationship-enrichment counseling with individual couples. Journal of Counseling Psychology, 44, 381-389. doi:10.1037/0022-0167.44.4.381

68.

Yovetich

N. A.

Rusbult

C. E.

(1994). Accommodative behavior in close relationships: Exploring transformation of motivation. Journal of Experimental Social Psychology, 30, 138-164. doi:10.1006/jesp.1994.1007