When couples experienced less negative communication than usual, they were also more satisfied with their relationship; positive communication was rarely associated with relationship satisfaction at the within-person level

Within-Couple Associations Between Communication and Relationship Satisfaction Over Time. Matthew D. Johnson et al. Personality and Social Psychology Bulletin, May 24, 2021. https://doi.org/10.1177/01461672211016920

Abstract: Relationship science contends that the quality of couples’ communication predicts relationship satisfaction over time. Most studies testing these links have examined between-person associations, yet couple dynamics are also theorized at the within-person level: For a given couple, worsened communication is presumed to predict deteriorations in future relationship satisfaction. We examined within-couple associations between satisfaction and communication in three longitudinal studies. Across studies, there were some lagged within-person links between deviations in negative communication to future changes in satisfaction (and vice versa). But the most robust finding was for concurrent within-person associations between negative communication and satisfaction: At times when couples experienced less negative communication than usual, they were also more satisfied with their relationship than was typical. Positive communication was rarely associated with relationship satisfaction at the within-person level. These findings indicate that within-person changes in negative communication primarily covary with, rather than predict, relationship satisfaction.

Keywords: couples, communication, longitudinal, relationship satisfaction

Discussion

Do within-couple changes in communication predict future changes in relationship satisfaction? This notion has been foundational to theories, empirical studies, and intervention protocols aimed at understanding and improving couple relations, but the literature has yet to adequately test this fundamental question. This study tested this question by drawing on data from three longitudinal studies of mixed-sex couples comprising community-based and national samples that used observational and self-report measurements at 4-month and annual assessment intervals. The pattern of results across these three studies supports two overarching conclusions regarding the within-person associations between couple communication and relationship satisfaction. First, there was inconsistent evidence for associations between within-person deviations in communication on lagged changes in satisfaction (or for satisfaction on lagged changes in communication). Second, there was consistent evidence for concurrent within-person associations between negative communication and satisfaction (but not positive communication and satisfaction), such that couples were more satisfied than normal at times when they also engaged in less negative communication than was typical.

Lagged Within-Couple Communication and Relationship Satisfaction Associations

We did not find consistent evidence that within-couple changes in communication prompted subsequent deviations in relationship satisfaction. Across studies, intraindividual deviations in positive communication were rarely a significant predictor of future changes in relationship satisfaction, indicating that increases in the extent to which partners expressed themselves and tried to understand their partner better were not associated with subsequent improvements in their relationship satisfaction. Changes in negative communication sometimes predicted worsening relationship satisfaction at the next measurement occasion, but no specific pattern was evident in every study. That is, despite the fact that each study found at least one set of linkages between negative communication and satisfaction in the direction expected by decades of theory, this pattern was not reliably observed and differed in specifics across the three studies. In Study 1, women’s self-reported negative communication predicted their own satisfaction. In Study 2, men’s self-reported negative communication predicted their own and their partner’s satisfaction. In Study 3, women’s self-reported negative communication predicted their partner’s satisfaction. Thus, although there were significant longitudinal within-person links between negative communication to future relationship satisfaction, the lack of cross-study support for any specific pathway casts doubt on the notion that communication is a robust, significant predictor of future changes in relationship satisfaction within couples.

Results were similar for the relationship satisfaction-to-communication within-person lagged links. As with the positive communication-to-satisfaction lagged associations, there was very little support for significant lagged satisfaction-to-positive communication linkages. There was relatively more support for significant lagged satisfaction-to-negative communication linkages; notably, there was a similar level of empirical support for the pathways from within-person changes in relationship satisfaction to future reductions in negative communication as there was from negative communication to future decreases in relationship satisfaction. Once again, however, no consistent longitudinal links emerged across studies. Broadly, such inconsistency in the within-person lagged communication/satisfaction association casts doubt on the robustness of these effects. Insofar as within-person fluctuations in couple communication and satisfaction might be linked, it is most likely these associations are due to changes in negative communication rather than positive communication. Furthermore, it is just as likely that fluctuations in satisfaction lead to deviations in negative communication as it is for negative communication to influence later relationship satisfaction.

Our ALT-SR analyses also considered between-person associations in within-person change trajectories of communication and relationship satisfaction (captured in their slopes). These analyses are not detailed in the article, due to length constraints, but are presented in the Supplemental Material (see Supplemental Table 17). These results are relevant to the cross-lagged within-person associations presented here because the pattern of longitudinal results at the between-person level mirrors those at the within-person level. Intercepts of communication and relationship satisfaction were inconsistently associated with the slope of the other construct and slope-to-slope associations were rare. Furthermore, several of the longitudinal between-person associations were in a counterintuitive direction (e.g., higher initial satisfaction was associated with a more gradual decrease in negative communication over time). Such counterintuitive between-couple findings are not without precedent (Gottman & Krokoff, 1989; Karney & Bradbury, 1997) and cast further doubt on the robustness of lagged associations between communication and relationship satisfaction.

Concurrent Within-Couple Communication and Relationship Satisfaction Associations

The concurrent within-couple associations between communication and relationship satisfaction paint a clearer picture. Once again, positive communication was not robustly linked with relationship satisfaction across studies. Consistent positive communication/satisfaction associations emerged in Study 3, but the coefficients were small in magnitude (rs from .05 to .07). The concurrent links between within-couple deviations in negative communication and relationship satisfaction, however, were robust across studies: Upward deviations in one partner’s negative communication generally coincided with intraindividual reductions in relationship satisfaction for oneself and the partner. This pattern was evident in 13 out of 16 total models tested across the studies, with the primary inconsistent result coming from men’s observed negative communication not being associated with their own or their female partner’s satisfaction. These results indicate that deviations in one partner’s negative communication are likely accompanied by concurrent changes in both partners’ relationship satisfaction, while within-couple deviations in positive communication unfold independent of relationship satisfaction. These findings extend a large body of research on between-couple associations between communication and relationship satisfaction (e.g., Woodin, 2011) and contribute additional evidence regarding concurrent within-person links (e.g., Nguyen et al., 2020) by indicating that not only do more satisfied couples communicate more positively and less negatively than dissatisfied couples during conflict, but that couples are most satisfied than normal at times when they are communicating less negatively (though not more positively).

In considering these findings, it is important to contrast the results in our analysis with those of Nguyen et al. (2020) who published the only other study to our knowledge to examine cross-sectional within-person links between relationship satisfaction and communication. Nguyen et al. found fluctuations in positive and negative (for women) communication were linked with concurrent deviations in relationship satisfaction, whereas this study found the most robust evidence in support of negative communication/satisfaction links for both partners. One possibility for the discrepant findings is that communication in this study was assessed specific to conflict situations, raising the possibility that the negative valence of such interactions may explain why negative communication exhibited more robust links with relationship satisfaction than positive communication. Nguyen et al.’s observed communication measure was coded on the basis of one problem-solving task and two support tasks. Given that prior research demonstrated that positive communication exerts more influence than negative communication in the context of positively valenced interactions, such as when partners share positive events with one another (Gable et al., 2006), it is possible that the inclusion of supportive interactions in the assessment of communication accounts for the more robust positive communication/satisfaction association (and lack of a main negative communication/relationship satisfaction effect for men). It is important for future research to continue exploring within-person communication-satisfaction associations in other contexts.

Limitations, Future Directions, and Implications

The findings presented here must be interpreted in light of limitations across these studies. First, as just mentioned, our focus on conflict and problem-solving communication leaves open the possibility that broader communication patterns or communication in other contexts would have different associations with satisfaction. Second, Studies 1 and 3 did not have optimal measurement. Study 1 did not assess positive communication and Study 3 relied on shortened measures of all constructs. These studies drew from large national samples of couples, as well, necessitating the use of self-reported assessment. The inclusion of Study 2 with robust measurement (observed and self-reported) from a less diverse sample was particularly valuable in this regard. Third, the time lags in each study (4 months in Study 1 and 1 year in Studies 2 and 3) cannot address the possibility that there may be more consistent short-term associations between couple communication and relationship satisfaction.¹ Along these lines, all self-reports of communication and relationship satisfaction were assessed with trait-like wording, asking participants to report how they typically communicate during conflict situations and their overall relationship satisfaction. State-like measurement focused on communication and satisfaction at a specific point in time (as was the case with the observational measurement in Study 2) coupled with daily or weekly data collection protocols would be valuable to determine whether linkages between communication and satisfaction might exist over shorter lags than used here.

With these limitations in mind, the results from this research have theoretical and empirical implications. Although couple communication has long been theorized as a central predictor of relationship satisfaction (Bandura, 1977; Gottman, 1979; Markman, 1979; Wills et al., 1974), scholars have questioned the robustness of the communication to relationship satisfaction pathway (Johnson & Bradbury, 2015; Karney & Bradbury, 2020; Lavner et al., 2016). Findings from this study provide novel longitudinal within-person evidence that underscore such concerns; even though some significant cross-lagged links were evident in the analyses, within-couple deviations in communication were not as robustly associated with future relationship satisfaction as would be anticipated based on the behavioral models of intimate relations. This study did reveal, however, robust concurrent within-person links between negative communication and relationship satisfaction: Men’s and women’s relationship satisfaction was higher than average at times when negative communication was lower than average.

These results suggest that the linkage between negative communication and satisfaction may be better conceptualized as one consisting predominately of covariation rather than prediction of change. Such a conclusion aligns with findings from a large machine learning study that examined a wide swath of individual and relational (including conflict) between-person predictors of relationship satisfaction in 43 longitudinal couple studies (Joel et al., 2020). Cross-sectionally, these analyses accounted for up to 46% of the variation in relationship satisfaction, but the longitudinal prediction of future changes in satisfaction proved elusive: Only 5% of the variance in change in satisfaction was predicted in any analysis. Together, these findings suggest that relationship science has been more successful in understanding what predicts how satisfied people feel with their relationship in the moment than in the future.

Given that all relationship-relevant processes play out at some level via verbal or nonverbal communication (Heyman, 2001), it is critical for the field to better articulate exactly how communication plays a role in couples’ relationship satisfaction. We offer three suggestions in this regard. First, an increased focus on moderators of the within-person longitudinal linkages between communication and relationship satisfaction could articulate the conditions under which a deviation in a given couple’s communication patterns is likely to be more or less impactful on their relationship satisfaction. Much valuable research in this vein has focused on additional dimensions of couple communication, such as the directness of the communication (Overall et al., 2009) or severity of the problems being discussed (McNulty & Russell, 2010), that moderate the impact of conflict communication on satisfaction. Another possibility would be to focus on how intrapersonal explanations of partner communication behaviors (e.g., attributions) may buffer the impact of communication on relationship satisfaction (Bradbury & Fincham, 1990). Moderators outside of the couple relationship deserve attention as well, such as contextual influences and individual characteristics. For example, as mentioned, Nguyen et al.’s (2020) study found that within-person fluctuations in negative communication for men (and effective communication for both men and women) were associated with relationship satisfaction deviations only at times when they experienced higher than average stress levels. Continued research in these areas would be of considerable benefit to further refine understanding of the conditions in which couple communication is likely to be more consequential for relationship satisfaction.

Second, the robust concurrent within-person associations between negative communication and relationship satisfaction suggest that the focus on longitudinal associations between communication and satisfaction may be misplaced. Rather, reductions in negative communication may be a concurrent facet of increased relationship satisfaction, or conversely, increased relationship satisfaction may reflect reductions in negative communication. A shift toward emphasizing covariation raises interesting questions for future research about how these variables come to be linked, including whether couples do in fact form judgments about their relationship on the basis of their communication in the manner long suggested by behavioral theory (Bandura, 1977; Gottman, 1979; Karney & Bradbury, 1995; Koerner & Jacobson, 1994; Markman, 1979; Wills et al., 1974) or whether this covariation simply reflects a more global level of functioning such that at times couples are functioning well, they function well in multiple domains and at times they are functioning poorly, they function poorly in multiple domains.

Third, these findings suggest that theories of relationship development would benefit from increased attention to interpersonal factors other than problem-solving communication behaviors that might prove to be more robust predictors of change in relationship satisfaction over time. Along these lines, prior research has found links between positively valenced communication in nonconflict situations and relationship satisfaction. For example, recent work has demonstrated that within-person increases in perceived expressions of gratitude from one’s partner predicted an intraindividual reduction in future anxious attachment (Park et al., 2019). Insofar as reductions in anxious attachment might serve to increase relationship satisfaction, expressions of gratitude might be one interpersonal process that predicts changes in relationship satisfaction. Alternatively, prior research found that partner responses to positive events were a more robust between-person predictor of future relationship satisfaction than responses to negative events (Gable et al., 2006) and a large literature has found between-person associations between the provision of support during stressful circumstances and relationship satisfaction (Bodenmann, 2005). Future within-person examination of these and other nonconflict communication processes may prove useful to refine perspectives specifying how communication influences the development of relationship satisfaction.