The Effects of Perceived Procedural Justice on Conflict Management between Spouses, and the Mediating Role of Dyadic Adjustment

Interpersonal conflict is integral to intimate spousal relationships, and, while potentially destructive, it can also enhance the sharing of feelings and thoughts, and the feeling that one is loved by one's partner (Pietromonaco, Greenwood, and Barrett 2004). Its positive or negative effects rely, in part, on how intimate partners deal with it (e.g., Rahim 2001; Laursen and Hafen 2010).

Conflict researchers often evaluate conflict management approaches by using a dual‐concern model (e.g., Thomas 1976; Rahim 1983, 2001) that originated in the study of organizational conflicts, but that researchers have since applied to more varied contexts (e.g., Pistole 1989; Ben‐Ari and Hirshberg 2009; Nelson, Schechter, and Ben‐Ari 2014). The model (Rahim 1983) incorporates two independent concerns, concern for self and concern for the other party, and identifies five distinct conflict management styles:

• Dominating/competitive: a person high in concern for self and low in concern for others;

• Obliging/concession making: at the opposite end of the scale from dominating is a person with low concern for self and high concern for others;

• Compromising: people with moderate concern for both themselves and others fall into this category;

• Integrating: high concern for both themselves and others characterizes people in this category who take a creative, problem‐solving attitude and try to address as many of each party's interests as possible; and

• Avoiding: not thinking about it; claiming it does not exist or matter; people in this category typically have low concern for both the interests of themselves and the other party.

Ali Kazemi (2007) characterized the three approaches that represent moderate to high levels of concern for the other—integrating, compromising, and obliging—as collaborative, because they reflect a person's readiness to satisfy the other party's needs; dominating and avoiding, he proposed, are non‐collaborative.

Studies have found correlations between how spouses deal with conflict and their relationship satisfaction, their mental health and well‐being, their physical health, and their chances of divorce (Gottman 1994; Karney and Bradbury 1995; Robles and Kiecolt‐Glaser 2003; Whitson and El‐Sheikh 2003; Whisman 2007; Segrin, Hanzal, and Domschke 2009). The results of these studies suggest that the antecedents of collaborative versus non‐collaborative conflict management approaches in intimate relationships can have significant impact on relationship health. This study contributes to our understanding of intimate partner conflict by examining the role that perceived procedural justice plays as an antecedent of conflict management behavior among spouses and by testing its direct and mediated effects.

Perceived justice refers to how individuals perceive decision making in their groups (states, communities, workplaces, schools, etc.), and its research has spanned many decades and different models (Tyler and Blader 2000, 2003). Originally, researchers concentrated on perceptions of distributive justice (Walster, Walster, and Berscheid 1978), which is the fairness of decision outcomes (e.g., fairness of resource allocation). Later, scholars recognized the significance of procedural justice, which are perceptions of the decision making process, how the decision was reached. Several studies have suggested that these perceptions affect people's attitudes even more than the decision outcomes (Tyler and Blader 2003).

The concept of perceived procedural justice (PPJ) first focused on the fairness and impartiality of the decision making process—for example, whether is it based on acceptable and impartial criteria (Thibaut and Walker 1978). Later, it expanded to include the relational aspects of the decision making process (Tyler and Blader 2003). Tom R. Tyler and E. Allan Lind (1992) proposed a comprehensive relational model of PPJ, comprising four dimensions that combine to create the perception that decision makers are behaving fairly:

1. Neutrality: the decision maker is perceived to be impartial and objective, considering facts and avoiding a whimsical process;¹

2. Trust: the decision maker is perceived as trustworthy, specifically, she shows consideration for the concerns of persons affected by the decision;

3. Standing: the decision maker is respectful and polite in her communications about the decision; and

4. Voice: the decision maker gives “voice” to others, in other words, she allows them to speak and listens to what they say (see also Tyler 1989; Van den Bos 2005; Van den Bos et al. 2013).²

Perceived procedural justice enhances individuals’ positive perceptions of their groups (i.e., families, organizations, etc.; Tyler 1989) and of their own status in them. Therefore, PPJ enhances individuals’ self‐worth, and their identification with, and commitment and loyalty to, their groups (Folger and Konovsky 1989; Tyler and Blader 2000, 2003; Terwel et al. 2010). Identification, commitment, and loyalty, in turn, induce individuals’ inclination to actively collaborate with decisions (De Cremer and Tyler 2007).

Studies have found correlations between PPJ and employees’ commitment, job satisfaction, productivity, helpful behaviors, reduced negative emotions, and reduced turnover (Brockner et al. 2007; Van Dijke, De Cremer, and Mayer 2010; Bal et al. 2011; Zhao, Peng, and Chen 2014), and with employees’ acceptance of unwelcome organizational changes (Kernan and Hanges 2002; Rubin 2009). During employee–manager conflict, PPJ positively correlated to the use of collaborative conflict management behaviors (Rahim, Magner, and Shapiro 2000; Spreitzer and Mishra 2000; Kazemi2007^,3; Giacomantonio, Pierro, and Kruglanski 2011). Rebecca Hollander‐Blumoff and Tom R. Tyler (2008) associated procedural justice perceptions with the inclination of law students to accept the outcomes of simulated dyadic dispute negotiations. Greg Berman and Emily Gold (2012) recommended that judges embrace PPJ principles to enhance criminal court communication. We recently studied the effects of PPJ on high school students during conflict with their teachers and found that PPJ correlated to fewer dominating and more obliging, compromising, and avoiding behaviors (Nelson et al. 2014). While the above findings concern organizational settings, Mark R. Fondacaro, Shelly L. Jackson, and Jennifer Luescher (2002) studied procedural justice in families, as perceived by adolescents, and found that it was inversely correlated with levels of conflict in the family.⁴

The literature on justice perceptions in intimate, romantic relationships has overwhelmingly focused on distributive justice (e.g., household division of labor; Kluwer, Heesink, and Van de Vliert 1997). Recently, scholars introduced PPJ into the study of intimate conflicts, pointing out their informal and highly interpersonal character. Process and relational aspects of decision making are no less or more important than decision outcomes in such relationships (Van Erp et al. 2011). The first studies in this line of research (Kluwer, Tumewu, and Van Den Bos 2009) tested whether men and women differed in their reactions to fair treatment during intimate conflict. In these studies, the perception of fairness was measured in several ways, including, for example, with questionnaires whose items essentially corresponded to Tyler and Lind's (1992) PPJ dimensions. Alternatively, they asked questions gauging respondents’ overall perceptions of whether their spouses treated them fairly during a conflict. Among other findings, Esther Kluwer, Maureen Tumewu, and Kees Van Den Bos (2009) found that PPJ (fair treatment) enhanced positive feelings and relationship satisfaction among women more than men. For men, the level of relationship commitment moderated the effects of PPJ; it increased positive affect and satisfaction only among highly committed male spouses.

More recently, Kim J.P.M. Van Erp and her colleagues (2011) studied perceptions of distributive and procedural justice and their associations with task‐related and personal conflict between expatriate employees and their accompanying spouses. They operationalized PPJ as a two‐dimensional construct, including the level of participation in decision making and the interpersonal treatment by the spouse (participation may correspond to the voice dimension, and interpersonal treatment to trust and standing). Among other findings, the study reported that caring and respectful interpersonal treatment, as perceived by accompanying spouses, reduced levels of personal conflict for both spouses.

Our current studies expand this line of research by testing how PPJ contributes to the choice or prediction of specific conflict management approaches during spousal conflict. We assume (probably much like previous researchers of PPJ in intimate relationships) that expressions and therefore perceptions of procedural justice are specific and distinct aspects of spousal communication that merit studying separately from other assessments of marital quality or communication. Perceived procedural justice specifically concerns processes of decision making in the relationship, and it points to concrete behaviors that enhance the sense that the spouse is fair and that decision making is participatory.

We embedded our research in the comprehensive PPJ model that Tyler and Lind (1992) proposed, and conducted two studies to test the association between spouses’ perceptions of PPJ and their choice of conflict management approaches. In the first study, we conducted an experiment to test how PPJ affected participants’ predictions about which conflict management style a fictional spouse would express. In the second study, we explored whether correlations between PPJ and conflict management approaches were mediated by the levels of dyadic adjustment, a measure of relationship health, in participants’ relationships with their own spouses.

The purpose of the first study was to investigate whether there is a causal link between PPJ and conflict management behaviors during spousal decision making. Based on the literature that established the associations between PPJ and conflict management among employees, school students, and so on, we expected to find similar results among spouses. We chose to conduct an experiment, because we hoped to determine whether PPJ influenced conflict management styles and not vice versa—for example, we would see a reverse causal relationship if a spouse who used dominating behaviors later justified such an approach as a response to his/her partner's procedurally unjust behavior. Therefore, we manipulated levels of PPJ during a hypothetical spousal decision making process.

Specifically, we manipulated the behavior of a spouse during the decision‐making process, so that participants could judge the degree of procedural justice he/she displayed (in the relational model, presented above, the fairness of the process is assessed by dimensions of the decision maker's behavior). We then tested how those PPJ assessments affected participants’ prediction of different conflict management approaches in a mutual decision making process.

Our Study One hypothesis is as follows:

We recruited participants from classes at Bar‐Ilan University, from various workplaces, and via Facebook and electronic mail. Two hundred and forty‐one Israelis (82 men and 159 women), aged 21–67 (M = 38.25, SD = 11.65) participated in the study. The sample included married people or people in committed non‐marriage relationships, but each participated as an individual. Roughly two‐thirds (64 percent) of the participants had children.

We used scenarios to manipulate levels of PPJ. We relied on Tumewu's (2009) scenario, which depicts a conflict over weekend plans. She manipulated the “voice” component of PPJ only, but because this is the first study to test PPJ's influences on conflict management orientations in the context of spousal conflict, we chose to follow Tyler and Lind's (1992) full relational model and operationalize PPJ according to all the dimensions of that model, to enhance its construct validity. Therefore, we wrote additional scenarios creating an experimental design of two (high/low) by four (neutrality, trust, standing, and voice), eight scenarios in total. Each participant received only one scenario, randomly assigned to him/her, representing one experimental condition. Group sizes ranged between twenty‐nine and thirty‐three.

All the scenarios were in Hebrew, which we have translated for this article. They all began with this description: “Dana is in conflict with her spouse Ofer. They need to decide what to do over the weekend. Ofer wants to meet with a few of their friends, and Dana wants an intimate weekend, just the two of them. Additionally, Dana had a very busy work week and she feels in need of some rest. When Dana approached Ofer and tried to tell him why she wanted time alone with him…”

The second half of the scenario represents the manipulation, accordingly:

• High neutrality: “… he explained to her his reason for wanting to spend time with friends. He thought that because they were new in their town, they should spend time and make an effort to enhance their social life. He explained that a group of their new friends were planning a birthday for one of the group, so it was a great opportunity for them to participate and fit in. He also explained to Dana that because they received a personal invitation, he felt very uncomfortable not attending.”

• Low neutrality: “… he told her that he felt like hanging out with their friends this weekend, and that he didn't think he always had to provide her with rational explanations for what he felt like doing.”

• High trust: “… he said that he saw her side and understood why she would want rest after such a busy week. He said that their relationship and their time alone was also very important to him, but because they were new in town, and didn't really establish a social life there, it would be good for both of them to spend time with their new friends this weekend. He really wanted to answer her needs and make her happy, but he felt that time with their friends would do them both good.”

• Low trust: “… he said that he actually really felt like hanging out with friends and having some fun this weekend. He also said that the most important thing to him right now was not to let down the friends and become known as someone who stands his friends up.”⁵

• High standing: “… he first made her a cup of coffee and suggested they sit together on the porch and discuss it. He listened to her pleasantly, and only then explained that in his opinion, because they were new in town and didn't have many local friends yet, it would be right for both of them to spend the weekend with their new friends.”

• Low standing: “… he answered impatiently while continuing to play with his smart phone. He argued that her real reason for not wanting to spend time with their friends was that she was lazy and unfriendly, so he doesn't see any real reason for turning down the friends’ offer to do something together. Before he finished these accusations, his phone rang and he took the call, very pleasantly answering the call from a friend.”

• High voice: “… he listened to her and tried to understand why it was so important to her. Only after he heard everything she had to say, did he present his side and say that because they were new in town and didn't have many local friends yet, he thought it would be right for both of them to spend the weekend with their new friends.”

• Low voice: “… he didn't let her explain herself and express her position. He kept saying that he wanted to spend the weekend with friends and to him, that was not negotiable.”

• After each participant read his or her assigned scenario, they responded to the following questionnaires.

To assess perceived levels of procedural justice in the scenarios (manipulation check) we used a set of four sub‐questionnaires (five to eight items each), representing the dimensions of PPJ identified above (neutrality, trust, standing, and voice). We based them on the questionnaires used in a study that examined high school students’ perceptions of their teachers’ procedural justice conducted by Noa Nelson, Dikla Shechter, and Rachel Ben‐ari (2014), which were Hebrew versions based on questionnaires from two previous studies (Tyler 1997; Tyler and Blader 2000). We changed the wording of the items to apply them to a spousal relationship and to the scenarios that our participants had read.

Participants rated their agreement with each statement on a six‐point Likert scale (1 = not true at all, 6 = very true), such that a higher score represents a greater perception of procedural justice. We tested the reliability of the statements using 110 participants with the following results (measured by Cronbach's alpha): neutrality (e.g., “Ofer was being balanced and relied on factual, relevant information before making a decision”) yielded alpha = 0.83; trust (e.g., “Ofer considered Dana's welfare when making a decision”) yielded alpha = 0.90; standing (e.g., “Ofer was treating Dana respectfully”) yielded alpha = 0.90; and voice (e.g., “Ofer gave Dana a chance to express her wishes”) yielded alpha = 0.83.

To assess conflict management styles, participants answered a Hebrew version of M. Afzalur Rahim's (1983; Hebrew translation by Ben‐ari, Feiner, and Shamir 2002) conflict management styles questionnaire. It includes five subscales, representing the five styles reviewed above (integrating, compromising, obliging, dominating, and avoiding). Together they consist of thirty‐two items, on a six‐point Likert scale (1 = completely untrue; 6 = very true), such that a higher score represents stronger endorsement of the approach. The wording of the items, typically used for respondents to report their own conflict management orientation, was changed so that participants could report what they thought Dana (the female spouse in the scenario) would do.

Following are the subscales’ reliabilities (measured by Cronbach's alpha): integrating (e.g., “Dana would try to explicitly express all considerations, so that the issues would be resolved in the best way”) yielded alpha = 0.93; compromising (e.g., “Dana would give something up, to receive something in return”) yielded alpha = 0.78; obliging (e.g., “Dana would oblige the wishes of her spouse”) yielded alpha = 0.84; dominating (e.g., “Dana would use her power to win the argument”) yielded alpha = 0.81; avoiding (e.g., “Dana would try to avoid difference of opinion with her spouse”) yielded alpha = 0.81.

We verified that participants perceived high or low levels of PPJ as intended in the scenarios, by comparing their scores on the PPJ questionnaires. We conducted a multivariate analysis of variation (MANOVA) in a two (high/low PPJ) by four (PPJ dimension) design, which yielded a significant and strong overall difference between high and low PPJ scenarios (F(1,239) = 422.78, p < 0.001, Eta² = 0.64). Table One presents descriptive statistics and differences between high and low PPJ, separately by dimension.

As can be seen in Table One, the manipulation of high versus low levels of PPJ worked as intended in all scenarios, yielding a significant and large difference in participants’ perceptions of each of its dimensions (see Cohen 1988 for a discussion of effect size). This means, for example, that participants who read the high neutrality scenario rated Ofer significantly higher on the neutrality scale, compared with participants who read the low neutrality scenario, and so on.

We note that we tested how each scenario affected the perceptions of all four dimensions of PPJ. We did this because PPJ dimensions had previously been highly correlated (former studies also averaged them into a unified PPJ score: Tyler 1997; Nelson et al. 2014).

We found indeed that these dimensions were hard to differentiate. Each high‐PPJ scenario enhanced all four PPJ dimensions, compared with its low‐PPJ counter‐scenario, and the dimensions were highly correlated (r = 0.60 to r = 0.82). This might suggest that the dimensions are truly intertwined, as found before. Alternatively, it might suggest a methodological issue, namely, that the scenarios we wrote were not adequately differentiated. To strengthen the former interpretation, we report correlations among the four dimensions in the second study. For the purposes of the current experiment, however, because the dimensions were highly correlated, and high versus low PPJ was correctly and strongly manipulated in all scenarios, we pooled the high‐PPJ experiment groups and their low‐PPJ counterparts, to create a unified PPJ variable with only two values, high and low.

To test Hypothesis One—that having higher perceptions of procedural justice would affect the prediction of collaborative behaviors and reduce the prediction of non‐collaborative behaviors—we conducted a MANOVA in which PPJ (high vs. low) was the independent variable and the five conflict management styles were the dependent variables. It yielded a significant overall medium effect for PPJ (F(5,235) = 3.13, p < 0.01, Eta² = 0.062). In Table Two we present the results of additional analyses of variance that we conducted separately for each dependent variable.

As Table Two shows, Hypothesis One was mostly supported. Participants who perceived higher levels of procedural justice were more likely to predict integrating behavior (this effect was the strongest); they were also somewhat more likely to predict compromising behavior and marginally more likely to predict obliging behavior. Perceived procedural justice also somewhat reduced predictions of avoiding behavior. Contrary to the hypothesis, PPJ did not affect participants’ predictions of dominating behavior.

The means in the table also show that the study's participants predicted collaborative behaviors (integrating, compromising and obliging) in an intimate, spousal conflict more than they predicted the non‐collaborative ones (dominating and avoiding).

The purpose of the first study was to establish a causal link between PPJ during spousal conflicted decision making and conflict management behaviors. We applied a method in which participants read scenarios and reported how they expected the person in the scenario to behave. Presumably, this method allowed them to project their own preferences without being as explicitly aware that they were being assessed, thus lowering social desirability motivations and enhancing honesty on their part. To strengthen the validity of the test, we recruited only people involved in a long‐term relationship, because we thought that their answers would reflect relevant personal experience when compared with people who are not in such a relationship.

The indirect expression of one's approach through projection, however, may fail to fully gauge the effects of PPJ on actual conflict management behavior in real life. This could explain why the overall effect was moderate and for most styles (except integrating) it was small (Cohen 1988). Moreover, in the scenario's experiment, participants were “endorsing” conflict management behaviors on behalf of a female character (Dana), and their choices could have been affected by gender stereotypes about prevalent or even acceptable female conflict management (see Nelson et al. 2015). To overcome these limitations, in the second study we asked respondents about their own relationships.

The results of the first study suggest that PPJ affects attitudes toward behaviors for managing spousal conflict, so that collaborative approaches, integrating in particular, are enhanced and non‐collaborative ones are partially inhibited. We will further discuss these findings in the general discussion, incorporating insights from the second study.

We conducted the second study with the following purposes in mind:

1. to enhance our findings’ validity, we tested PPJ's contributions to conflict management styles in a more naturalistic setting, asking respondents to report on their approaches for managing conflict in their own relationships; and

2. to expand our initial knowledge about PPJ's contributions to intimate conflict management styles by exploring its mechanism, that is what variables might specifically mediate (e.g., explain) the correlations between PPJ and conflict management behaviors between spouses.⁶

Regarding the first purpose, we tested for a direct correlation between participants’ perceptions of procedural justice in their spousal relationship and their conflict management approaches. We expected that PPJ would positively correlate to collaborative styles and negatively correlate to non‐collaborative styles. Although the results of our first study weakened our expectations of a correlation between PPJ and dominating (and obliging was only marginally affected), we nevertheless tested all five styles in the second study. Not only is it premature to assume that these behaviors are not correlated to spousal PPJ based on a single study, but also Study Two differs from Study One in that its respondents directly reported their own approaches during conflicts with their spouses.

As for our second purpose, we wanted to shed light on the psychological mechanism that connects spousal PPJ to conflict management styles. According to the models developed by Tom Tyler, E. Allan Lind, and Steven L. Blader (Lind and Tyler 1988; Tyler and Lind 1992; Tyler and Blader 2003), procedural justice behaviors suggest to individuals that they are being treated fairly, respectfully, and with care. Such perceptions enhance their self‐worth, and feelings of belonging and having a strong position in a group or a relationship, and also strengthen the individual's commitment and loyalty toward and identification with the fair and caring person or group. Belongingness, commitment, and identification, in turn, enhance collaboration during conflict.

“Dyadic adjustment” (DA) is a term that describes a sense of dyadic worth (corresponds to the group's worth), that is, the feelings that both members of a dyad have of belongingness and commitment toward and identification with the pair/couple. The DA construct encompasses the level of consensus between intimate partners, their dyadic cohesiveness, their satisfaction with their relationship, and the incidence of expressions of affection within the relationship (Spanier 1976; see also Montesi et al. 2010). We propose a mediation model in which PPJ positively correlates to dyadic adjustment, which in turn, positively correlates to collaborative strategies and negatively to non‐collaborative strategies.

As we noted earlier, some researchers have previously correlated PPJ to DA, reporting that spousal PPJ enhanced affective feelings and relationship satisfaction among women and highly committed men (Kluwer et al. 2009). Other studies, which did not directly address PPJ but somewhat similar qualities, such as mutual respect, integrity, and respectful communication, found that these qualities enhanced love, intimacy, and relationship satisfaction (Fehr 1988; Stanley, Markman, and Whitton 2002; Tumewu 2009) and support the expectation that PPJ would positively correlate to DA.

Dyadic adjustment, in turn, is expected to correlate to specific conflict management approaches. Saadia Dildar, Aisha Sitwat, and Sumaira Yasin (2013) tested relationship satisfaction (one dimension of DA) and conflict management among Pakistani couples, and found that dissatisfaction correlated to increased use of non‐collaborative strategies and decreased use or absence of collaborative ones. Others have shown that relationship satisfaction affects whether people interpret their spouses’ behaviors positively or negatively regardless of the objective content of the behavior (Weiss 1980; Flora and Segrin 2000; Hawkins, Carrère, and Gottman 2002).

When dyadic adjustment is high and spouses interpret behaviors positively, we expect that they will collaborate more. More generally, marital adjustment relates to lower psychological distress and the negative emotional responses distress elicits (Li, Robustelli, and Whisman 2015), so it indirectly enhances the ability to negotiate conflicting interests collaboratively and effectively (Nelson, Shacham, and Ben‐ari 2016).

Taken together, PPJ models and studies on DA dimensions and spousal conflict support our proposed mediation model, in which the perception that a spouse is behaving justly would positively correlate to DA, and DA would positively correlate to collaborative conflict management behaviors and negatively correlate to non‐collaborative ones. We expected that DA would explain, at least partially, PPJ's correlation with conflict management approaches. Our hypotheses, which we articulated according to the requirements of mediation testing (see Baron and Kenny 1986), follow:

We recruited participants in this study as we did in Study One: by recruiting university students, in workplaces, and via Facebook and electronic mail. One hundred and sixty participants (115 women and 45 men), aged 25–71 years (M = 42.83, SD = 12.6), participated in the study. All were either married or in committed long‐term heterosexual relationships. Most participants had children, only 16 percent were childless.

Some respondents received questionnaires in paper form on campus or at various workplaces. Others answered an electronic version via the Internet. In both cases, we explicitly stated that the questionnaires were anonymous. To enhance the sense of anonymity, respondents who were physically handed paper‐and‐pen questionnaires did not answer them in front of the researcher, but mailed them in later. The instructions stated that everyone is sometimes involved in conflicts and some of those are with one's spouse. We encouraged respondents to answer as honestly as possible and give answers that describe their spousal relationship in general.

This set of four sub‐questionnaires is similar to the set used in Study One, except that the wording of items was minimally changed to allow respondents to report on PPJ in their spouses’ behaviors and decisions. Reliabilities (measured by Cronbach's alpha) of the questionnaires were as follow: neutrality (e.g., “my spouse explains his/her decisions to me”) yielded alpha = 0.79; trust (e.g., “my spouse is thoughtful of my needs”) yielded alpha = 0.88; standing (e.g., “even when my spouse is angry, she treats me respectfully”) yielded alpha = 0.81; and voice (e.g., “my spouse makes a point of hearing my opinion before making important decisions”) yielded alpha = 0.79. The correlations among PPJ's elements in the current study were high (r = 0.55–0.76) and, because previous studies used an average total PPJ score (Tyler, Degoey, and Smith 1996; Tyler 1997; Nelson et al. 2014), we also computed a PPJ score based on the entire set (alpha = 0.93).

This questionnaire, developed by Graham B. Spanier (1976), has been extensively used to assess intimate relationship quality. It comprises thirty‐two statements on differing Likert scales (from two‐point scales to seven‐point scales), and the scoring method gives more weight items with a larger range of answers (see Spanier 1976 for details). In all items, a higher score indicates higher dyadic adjustment. Spanier's (1976) original factor analysis yielded four dimensions:

1. dyadic consensus, which is the level of agreement on important issues (e.g., “to what extent do you and your spouse agree on religious matters?”);

2. cohesiveness, which assesses how often partners spend time together and share activities (e.g., “How often do you and your spouse exchange interesting ideas?”);

3. satisfaction, which is their level of satisfaction and sense of commitment to continuing the relationship successfully (e.g., “how often do you consider or discuss divorce?”); and

4. affection expressions (e.g., “Do you kiss your spouse?”).

Spanier (1976) reported high internal consistencies for each factor and for the overall questionnaire. Naomi Zoran (1995) translated the DAS into Hebrew and used it to yield an overall score (alpha = 0.92). In the current study we followed her and other researchers (Shavit 2009; Lomar 2011), and produced a total DA score based on all items (alpha = 0.92).

This questionnaire (Rahim 1983) is similar to the one used in Study One, except that the wording of items was changed to allow respondents to report on their own responses to their own spouses. It yielded the following reliabilities (measured by Cronbach's alpha): integrating (e.g., “I'll try to discuss the matter with my spouse, to find solutions that will answer both our needs”) yielded alpha = 0.91; compromising (e.g., “I'll give something up to receive something in return”) yielded alpha = 0.74; obliging (e.g., “I would oblige the wishes of my spouse”) yielded alpha = 0.72; dominating (e.g., “I'll use my power to win”) yielded alpha = 0.79; avoiding (e.g., “I'll try to avoid differences of opinion with my spouse”) yielded alpha = 0.81.

We note that participants in the second study, who were asked about their own relationships, reported lower usage of non‐collaborative strategies compared with collaborative ones (see Table Three), which raises the question of whether social desirability bias played some part in their responses. But we received similar results in Study One, which was specifically designed to reduce the impact; this suggests that social desirability bias is not a significant factor.

In constructing the second study, we acknowledged that participants’ predictions in the first study could have reflected gender stereotypes, because participants had predicted conflict management approaches on behalf of a female scenario character. In the current study, both women and men reported their own approaches. We conducted a MANOVA to test for gender differences and it yielded a small but significant overall difference (F(5,154) = 2.68, p < 0.05, partial eta‐squared = 0.08). Additional analyses of variance, however, indicated that men and women only significantly differed in obliging (F(1) = 3.03, p < 0.05, partial eta‐squared = 0.05), and that contrary to gender stereotypes, men's self‐ratings of obliging (M = 4.77, SD = 0.61) were somewhat higher than women's (M = 4.47, SD = 0.62). In other words, the preference for collaborative over non‐collaborative styles, found in the first study and again here seems to be truly characteristic of conflict between spouses, among both genders.

We also note that, similar to Study One (in which manipulating one PPJ dimension also affected the other three) PPJ's four dimensions in the current study were not well differentiated. The correlations between them were all significant and high, ranging between r = 0.55 and r = 0.76. These repetitive findings suggest that PPJ dimensions are highly correlated and that their correlations in Study One were not an artifact of manipulation materials. Given PPJ dimensions’ correlations to each other, we used an average PPJ score as an independent variable in the following hypotheses tests.

To test Hypotheses Two(a) through Two(c), we computed Pearson correlations among the research variables. Table Four presents the correlations of PPJ [Hypothesis Two(a)] and of DA [Hypothesis Two(b)] with conflict management strategies.

The results support part of Hypothesis Two(a)—that PPJ would correlate positively to collaborative strategies—but it did not support the second part; we found no negative correlation between PPJ and non‐collaborative strategies. As Table Four shows, PPJ's correlation with integrating was particularly strong (0.54) and only moderate with obliging.

Hypothesis Two(b)—that DA would correlate positively to collaborative strategies and negatively to non‐collaborative ones—was almost fully supported, except for its lack of correlation with avoiding. Dyadic adjustment's positive correlations to the collaborative strategies were stronger than its negative correlation to dominating (see Table Four), and again the correlation with integrating was particularly strong (0.61).

As for Hypothesis Two(c), the positive correlation between PPJ and DA was r = 0.66 (p < 0.001), strongly supporting it.

To test Hypothesis Two(d)—that DA would mediate the associations of PPJ with conflict management strategies—we conducted regression analyses, in each of which the predictors were PPJ and DA, and the predicted variable was one of the styles. We entered PPJ in the first step and DA in the second step of the regression. This method allowed us to test these predictors’ unique contributions to the variance in styles, and specifically to test indications for mediation, as explained below. Table Five summarizes the results of these analyses.

The regression analyses suggest that perceived procedural justice and dyadic adjustment both contributed positively to participants' choice of integrating behaviors (see Table Five), but PPJ's contribution weakened in the second step, when DA was added, indicating that DA partially mediated PPJ's contribution. A Sobel test verified this partial mediation effect (Z = 4.95, p < 0.001). A similar pattern is presented in Table Five for the positive contributions of PPJ and DA to compromising, and again the partial mediation was verified by a Sobel test (Z = 2.6, p < 0.01). As for obliging, PPJ contributed to it in the first step, but its contribution became insignificant (and almost zero) when DA's positive contribution was added in the second step. Such a pattern indicates full mediation by DA, which was verified by a Sobel test (Z = 4.12, p < 0.001). In other words, PPJ does not contribute directly to obliging at all; it predicts DA, and the latter predicts obliging.

Results support Hypothesis Two(d) with regard to collaborative approaches; DA mediated, partially or fully, PPJ's associations with participants’ inclination to collaborate with their spouses during conflict. Regarding the non‐collaborative strategies, the mediation hypothesis (Hypotheses Two(d)) was not supported because PPJ itself did not contribute to either dominating or avoiding. Dyadic adjustment did, however, contribute negatively to dominating: respondents who reported higher rates of dyadic adjustment were less likely to take a competitive, dominant attitude during conflict with their spouses.

To summarize, the perception that one's spouse is neutral, caring, respectful, engaged, and encouraging in decision making correlates positively to inclinations to integrative and compromising behavior, and, to a lesser extent, obliging behavior. Perceived procedural justice's association with obliging, however, was fully mediated by dyadic adjustment, which also partially mediated its association with integrating and compromising. In other words, the perception of procedural justice on the part of one's spouse, is related to higher dyadic adjustment (satisfaction, cohesiveness, affection, and consensus), and that, in turn, explains the inclination to oblige one's spouse, and partially explains the inclination to integrate and compromise.

Perceived procedural justice did not contribute at all to non‐collaborative behaviors, but we note that overall our participants reported less use of non‐collaborative behaviors during conflict with their spouses, compared with collaborative ones. These findings suggest that in intimate relationships, PPJ correlates to collaboration during conflict, but the absence of PPJ does not correlate to the less adaptive and less endorsed non‐collaborative behaviors.

Living together and sharing an intimate, meaningful relationship naturally leads also to conflict. While unavoidable, conflicts need not be destructive, and can in fact enhance intimacy and strengthen the relationship. These positive outcomes can be affected by spouses’ conflict management behaviors, and therefore it is important to study the antecedents of collaborative versus non‐collaborative conflict management in spousal relationships. The current studies contribute to this effort by exploring the role of perceived procedural justice.

In Study One, we manipulated levels of PPJ in fictional scenarios of spousal conflict and found that PPJ affected participants’ predictions of how the fictional spouse would behave, predictions that presumably reflected their own preferences. In scenarios in which a husband behaved in a procedurally just way, participants were more likely to predict that his wife would engage in integrative behavior in particular, but also in compromising, and to a lesser degree, obliging behavior. Perceived procedural justice also seemed to reduce avoiding, but did not affect dominating.

In Study Two, we questioned participants about their own relationships and conflicts, and tested a mediation model in which we examined whether PPJ's impact on conflict management could be explained by dyadic adjustment. We found that PPJ correlated positively to collaborative strategies, particularly integrating, and that DA partially mediated its correlation to integrating and compromising, and fully mediated its correlation to obliging. Results also showed that PPJ did not correlate to the non‐collaborative dominating or avoiding approaches.

Additionally, in both studies, participants scored non‐collaborative styles lower than they did collaborative ones. In other words, they predicted and reported using collaborative strategies more than non‐collaborative ones.

Taken together, these findings portray an encouraging picture of PPJ's role in intimate conflicts. We will first discuss its direct contributions to conflict management, and then DA's mediating role.

The perception that their spouse is behaving in a procedurally just manner encourages couples to adopt more collaborative behavioral approaches, especially integrating and compromising, during conflict. The impact of PPJ on integrating is notable, because it is a demanding behavior, requiring committed and creative discussions, to expose interests and attempt to answer both parties’ needs. But for the same reasons, it is also the style with the highest potential for satisfying parties and enhancing their relationship and their ability to resolve their future conflicts. This means that when people perceive their spouses as fair, impartial, caring, respectful, and participatory, they will work harder to resolve their differences and strengthen their relationship.

We further suggest that PPJ had only a marginal causal effect on obliging in Study One and no correlation to it in the second study (dyadic adjustment fully mediated the correlation) because when differences and decisions are discussed in a procedurally just way, spouses would be less likely to feel unable to express their needs and therefore less likely to choose obliging behaviors.

Finally, we find it interesting that the lack of PPJ did not enhance non‐collaborative strategies in spousal conflict, especially dominating (avoiding was enhanced by low PPJ in Study One). Our participants in both studies showed a preference for collaborative strategies over non‐collaborative ones. They seemed altogether to prefer adaptive and positive ways of managing intimate conflicts, and the above finding suggests that they do not resort to a selfish and potentially aggressive or destructive strategy, even when they perceived their spouse is not behaving in a procedurally just way. We note that this suggests that the absence of PPJ may have a different impact on intimate relationships than on organizational relationships; previous research has found that the lack of PPJ does correlate to dominating behavior in organizations (e.g., Spreitzer and Mishra 2000).

Dyadic adjustment partially mediated PPJ's correlations to integrating and compromising behaviors in spousal conflict. Participants who perceived their spouses as impartial, caring, respectful, and listening also reported higher levels of satisfaction, consensus, expressions of affection, and cohesiveness in their intimate relationships. That, in turn, seemed to partially explain their inclination to work together to find mutually satisfactory solutions, attend to both their needs, and to compromise with their spouses. Therefore, PPJ seems to positively affect spousal conflict management at least partially by enhancing the overall quality of the relationship, and motivating collaboration in the process. We note, however, that PPJ retained some direct correlations to integrating and compromising, when DA was accounted for. Therefore, it is possible that collaboration directly reciprocates PPJ (for more on reciprocity in conflict, see Putnam and Jones 1982; Brett, Shapiro, and Lytle 1998). Dyadic adjustment also fully explained PPJ's contribution to obliging, which in itself was null in the second study. This suggests that cohesiveness and affection, incidentally enhanced by PPJ, lead people to oblige their spouses in a selfless manner.

These studies did have several limitations. First, we examined the perceptions of individuals, not the dyad. Because spouses are interdependent (e.g., Montesi et al. 2010; Donato et al. 2015), their justice perceptions and conflict management behaviors are mutually affected. To address this, future research could explore PPJ and conflict management approaches in both spouses, applying the “Actor Partner Interdependence Model” (APIM), which accounts for individual, dyadic, and interactional effects (Kenny, Kashy, and Cook 2006; see also Van Erp et al. 2011; Bronstein et al. 2012). In addition, we did not control for relationship longevity. It seems possible that those couples whose relationships have survived for longer have evolved more effective conflict management behaviors, which could have associated impacts on PPJ and vice versa.

Another limitation affects only the first study, in which we used fictional scenarios to manipulate levels of PPJ. The protagonist in the scenarios was female, and we assessed participants’ conflict management styles through their prediction of her responses to her male spouse's behaviors. Gender‐related stereotypes of conflict behavior (Nelson et al. 2015) could have affected our participants’ responses. In that case, styles scores would not only reflect procedural justice, but could be confounded with what they think a female spouse is likely to do when handling her male partner. In the second study, both men and women considered their own relationships but we found that gender barely affected their self‐ratings, and when it did the effect was contrary to gender stereotypes (men reported more obliging behaviors than women). While this alleviated some of our concerns about the first study, we think that replicating our first study with a male protagonist would be worthwhile, as well as extending the exploration to same‐sex relationships.

Finally, both our studies were conducted in Israel, where the culture is fairly Western, and where social distance and power inequality are much less pronounced than in many other cultures (see Hofstede 2001). Previous studies suggested that cultural norms—for example, expectations for equality—tend to influence PPJ's significance (Brockner et al. 2001; Van den Bos et al. 2013), and that conflict management is culture‐dependent (Komarraju, Dollinger, and Lovell 2008). We believe, therefore, that conducting similar studies in other countries and cultures would be enlightening.

To our knowledge, these studies are the first to introduce and test the role of PPJ in the choice of conflict management approaches to spousal conflict, extending our understanding of the important role of justice—not just in social and political institutions but in families and interpersonal relationships. In addition, our findings about PPJ's positive contributions have implications for the practice of dyadic mediation as well as couples and family counseling. Procedurally just behavior can be taught and people can be made more sensitive to its consequences—in marriages and families as well as in workplaces, political bodies, and schools.