Closeness Discrepancies in Romantic Relationships: Implications for Relational Well-Being,Stability,and Mental Health

Abstract

Experiences of closeness in romantic relationships are associated with heightened levels of relational well-being and mental health. However, individuals differ in the degree of closeness they desire in their relationships. This study used the construct of inclusion of other in self (IOS) to analyze discrepancies between individuals’ actual and ideal levels of closeness in their relationships and the implications of these discrepancies for relational well-being and mental health. A longitudinal survey of partnered individuals demonstrated that optimal levels of relational well-being and mental health existed when individuals had minimal discrepancies between actual and ideal IOS, regardless of their actual levels of IOS. Individuals whose actual levels of IOS moved closer to their ideal levels over a 2-year period reported improved relational well-being and mental health. Individuals with little to no discrepancies between actual and ideal IOS were also less likely to break-up with their partners over time.

Keywords

Closeness Health Relationships Discrepancies Inclusion of Other in Self

Closeness, Health, and the Quality of Romantic Relationships

Close relationships with others—particularly romantic relationship partners—have a substantial impact on individuals’ health and well-being (Holt-Lunstad, Smith, & Layton, 2010; Kiecolt-Glaser, Gouin, & Hantsoo, 2010; Reis & Aron, 2008; Reis, Collins, & Berscheid, 2000; Robles & Kiecolt-Glaser, 2003). Feelings of closeness within romantic relationships play an important role in determining the quality of romantic relationships and partners’ mental health (Khaleque, 2004). Feeling close to a romantic partner is indicative of a more intimate, committed, and satisfying relationship (Brunell, Pilkington, & Webster, 2007; Hassebrauck & Fehr, 2002). Closeness in romantic relationships is further representative of the degree to which individuals include aspects of their partners within their own self-concept: inclusion of other in self (IOS; Agnew, Loving, Le, & Goodfriend, 2004; Aron, Aron, & Smollan, 1992; Aron, Aron, Tudor, & Nelson, 1991). Seeking self-expansion by including qualities of one’s partner in one’s self-concept (e.g., their identities, resources, experiences) is thought to produce beneficial outcomes at both the relational and individual levels (Aron, Norman, & Aron, 2001).

Research on IOS as an indicator of feelings of closeness in romantic relationships has demonstrated that greater inclusion of one’s partner in one’s self-concept is associated with a multitude of indicators of relational well-being, including greater relationship satisfaction and lower rates of dissolution over time (Le, Dove, Agnew, Korn, & Mutso, 2010; Tsapelas, Aron, & Orbuch, 2009). In addition, findings across a variety of cognitive and linguistic studies indicate that a greater sense of “we-ness” between partners within relationships is associated with increased relationship satisfaction and commitment (Agnew, Van Lange, Rusbult, & Langston, 1998; Fitzsimons & Kay, 2004; Frost, in press; Seider, Hirschberger, Nelson, & Levenson, 2009). Self-expansion through IOS may also result in beneficial relational outcomes by promoting positive interdependence between partners in the form of shared interests and mutuality of dependence (Rusbult & Van Lange, 2003). The majority of psychological research therefore indicates that closeness—or more overlap between one’s self and one’s partner—is a key marker of quality, healthy relationships.

Closeness Discrepancies and Their Potential Negative Effects

However, individuals vary in the amount of closeness they desire within their relationships (Aron et al., 2004; Fletcher, Simpson, Thomas, & Giles, 1999; Goodboy & Booth-Butterfield, 2009; Kashdan, Volkmann, Breen, & Han, 2007; Mashek, Le, Israel, & Aron, 2011; Mashek & Sherman, 2004). Recent advances in work on IOS have pointed to the phenomena of desiring less closeness in romantic relationships. Specifically, too much IOS may pose a threat to personal control and identity, and therefore cause someone to desire less closeness with an intimate partner (Aron et al., 2004; Mashek & Sherman, 2004). Similar problems arise when considering the phenomena of enmeshment in marital and romantic relationships. Increased experiences of closeness may be beneficial to relational well-being and mental health when they take the form of intimacy, caregiving, and nurturance, but detrimental when they take the form of intrusiveness and loss of a sense of self and independence (Green & Werner, 1996; Lewandowski, Nardone, & Raines, 2010; Werner, Green, Greenberg, Browne, & McKenna, 2001). Thus, feeling “too close” to one’s partner and the corresponding perceived loss of self may have a detrimental effect on relational well-being and mental health (Aron et al., 2004; Mashek & Sherman, 2004).

Taken together, the research reviewed previously suggests the utility of an understanding of feelings of closeness in romantic relationships that can account for individuals’ actual experiences of IOS in relation to their ideal experiences of IOS. In other words, individuals in romantic relationships who do not currently experience their ideal level of closeness with their partners may feel either “not close enough” or “too close” to their partners. Although the theoretical importance and existence of these discrepancies between actual and ideal experiences of IOS have been documented, the implications that IOS discrepancies actually have for the quality and stability of romantic relationships remain largely unexamined (Aron et al., 2004; Mashek & Sherman, 2004).

Research on individual differences in experiences of IOS discrepancies is further suggestive of the potential detrimental effects of IOS discrepancies on relational well-being. Attachment avoidance is associated with experiencing a discrepancy between actual and ideal IOS in relationships, whereas the lowest levels of attachment avoidance can be found among individuals who experience no actual–ideal IOS discrepancies (Mashek & Sherman, 2004). Avoidance in the form of fears of intimacy and perceived threats to independence (i.e., dismissive-avoidant attachment) may manifest in feeling “too close” to a partner, while desiring closeness yet anticipating rejection from a partner (i.e., fearful-avoidant attachment) may correspond to not feeling close enough to a partner (Bartholomew & Horowitz, 1991). Adult attachment theory positions attachment avoidance as detrimental to relationship quality and stability because attachment avoidance is characterized by a fear of intimacy, as well as an increased desire to maintain distance and independence at the expense of intimacy within the context of close relationships (see Hazan & Shaver, 1994, for a review). Empirical evidence exists for this connection as well: Attachment avoidance is negatively associated with multiple indicators of well-being in romantic and marital relationships (Birnbaum, Reis, Mikulincer, Gillath, & Orpaz, 2006; Brennan & Shaver, 1995; Butzer & Campbell, 2008). The co-occurrence of closeness discrepancies with attachment avoidance therefore suggests that actual–ideal IOS discrepancies will be indicative of negative relationship outcomes given the demonstrated deleterious effects of avoidance on relationship satisfaction and stability.

Social Cognitive and Motivational Perspectives on Closeness Discrepancies and Well-Being

Self-discrepancy theory (e.g., Higgins, 1987)—and its extension to discrepant self-concepts within dyads (e.g., Robins & Boldero, 2003)—provides some insight into how closeness discrepancies may have a deleterious impact on relational well-being, stability, and mental health. More specifically, one’s idealized level of closeness or IOS may serve as a cognitive guide to which one compares his or her actual experiences of closeness or IOS. Although not focused on closeness, discrepancies between actual and ideal selves can result in “dejection-related emotions” such as disappointment and dissatisfaction (Higgins, 1987). When idealized standards for characteristics of partners do not match perceived partner characteristics, people become dissatisfied with their relationships (Fletcher & Simpson, 2000).

Extending this line of reasoning to understanding the influence of closeness on relational well-being, it follows that discrepancies between actual and ideal levels of closeness may translate to similar outcomes at the relational level: decreased relationship satisfaction and commitment. Furthermore, following findings relating self-discrepancies to mental health (Heidrich & Powwattana, 2004; Scott & O’Hara, 1993), actual–ideal closeness discrepancies may result in negative mental health outcomes, such as depression, given the centrality of dejection-related emotions to depressive symptomatology (e.g., Radloff, 1977). Finally, just as individuals are motivated to reduce actual–ideal self-discrepancies (Higgins, 1987) and discrepancies between perceived and ideal partner characteristics (Fletcher & Simpson, 2000), the drive to bring one’s actual experience of closeness in line with one’s ideal experience of closeness may represent an energizing cognition that, when actualized over time, should result in positive trajectories in relational well-being and mental health. Conversely, failure to diminish actual–ideal closeness discrepancies may result in negative trajectories in relational well-being and mental health over time, and ultimately relationship dissolution.

Similar predictions about the effects of actual–ideal closeness discrepancies on relational well-being and mental health can be gained from self-determination theory (SDT; Ryan & Deci, 2000). Research using an SDT approach has demonstrated that relational well-being and mental health result from the satisfaction of fundamental psychological needs, including relatedness and autonomy, in addition to competency (e.g., Patrick, Knee, Canevello, & Lonsbary, 2007; Sheldon, Williams, & Joiner, 2003). Feeling less closeness than idealized may be indicative of an unmet need for relatedness in the context of a present partnership, whereas feeling more closeness than idealized may indicate, in a sense, over-relatedness, which poses a threat to actualization of the need for autonomy. Alignment of actual and ideal experiences of closeness may therefore represent an optimal balance between the needs for relatedness and autonomy as subjectively experienced by the individual. As a result of these threats to fundamental psychogenic needs for relatedness and autonomy, closeness discrepancies in romantic relationships may put individuals at risk for diminished well-being in both psychological and relational domains.

Beyond Actual Closeness: The Potential Additive Influence of Closeness Discrepancies on Relational Well-Being and Mental Health

Researchers have yet to understand the degree to which actual–ideal IOS discrepancies impact relational well-being above and beyond actual experiences of IOS alone. In other words, it is unclear what matters most for relational well-being: how close people actually feel to their partners, or whether people’s actual experiences of closeness in their relationships match how close they would like to feel to their partners. Closeness is a subjective psychological experience, and individuals have been shown to vary in how much closeness they idealize within their romantic relationships (Aron et al., 2004; Fletcher et al., 1999; Goodboy & Booth-Butterfield, 2009; Mashek & Sherman, 2004). Thus, a prescriptive model—one in which more closeness always results in better and healthier relationships—may not be the most effective theoretical, empirical, and clinical approach to understanding the role of closeness in determining relational well-being, stability, and mental health. Although there is evidence that, on average, closer relationships tend to be better relationships (Brunell et al., 2007; Hassebrauck & Fehr, 2002), understanding whether and how actual–ideal closeness discrepancies can predict relational well-being, stability, and mental health above and beyond actual closeness alone can potentially increase the explanatory power and applied relevance of psychological models of health and well-being in romantic relationships.

The Current Study: Aims and Hypotheses

The current study uses an analysis of closeness as IOS and seeks to extend and replicate social cognitive theory and evidence on actual–ideal discrepancies within the domain of closeness. Utilizing the construct of actual–ideal IOS discrepancies as developed in research on experiences of desiring less closeness (Aron et al., 2004; Mashek et al., 2011; Mashek & Sherman, 2004), the current study aims to examine the degree to which actual–ideal IOS discrepancies are associated with relational well-being, stability, and mental health above and beyond the actual experience of closeness alone. The implications of IOS and actual–ideal discrepancies for mental health in particular have yet to be examined. Furthermore—with increased empirical attention being paid to the ever-changing nature of romantic relationships (e.g., Berscheid, 2010)—it is imperative to understand how the attenuation or exacerbation of actual–ideal IOS discrepancies over time correspond to temporal changes in relational well-being, stability, and mental health. Accounting for dynamic experiences of actual–ideal IOS discrepancies and their resulting implications is a necessary next step for a psychological science that seeks to better its understanding of the experience of closeness, as well as improve its ability to promote and facilitate healthy, lasting relationships.

The current study hypothesized that discrepancies between actual and ideal levels of IOS would predict relational well-being outcomes (i.e., relationship satisfaction, commitment, and break-up thoughts) and mental health (i.e., depression) above and beyond individuals’ actual experiences of IOS alone. Furthermore, it was hypothesized that the association between IOS discrepancies and relational well-being and mental health outcomes would be nonlinear. Individuals with little to no discrepancies between their actual and ideal levels of IOS were expected to demonstrate optimal levels of relational well-being and mental health. Experiencing higher levels of IOS than desired (i.e., feeling “too close” with one’s partner) as well as lower levels of IOS than desired (i.e., feeling “not close enough” with one’s partner) were both hypothesized to be associated with decreased levels of relational well-being and mental health. Furthermore, changes in actual–ideal IOS discrepancies over time were hypothesized to be predictive of corresponding changes in relational well-being and mental health. It was hypothesized that individuals whose actual experiences of IOS with their partners moved closer to their ideal IOS over time would show improvement in relational well-being and mental health over time compared with individuals whose actual experiences of IOS with their partners moved further away from their ideal levels of IOS. Comparatively, little to no change in relational well-being and mental health was hypothesized among individuals with no change in actual–ideal IOS discrepancies over time. Finally, it was hypothesized that individuals with no actual–ideal IOS discrepancies would be less likely to break up with their partners over time than individuals who felt “too close” or “not close enough” to their partners.

Method

Sample

To test these hypotheses, data were obtained from the Lives and Relationships Study: an Internet-based longitudinal survey focused on understanding predictors of relational well-being and health among a diverse sample of individuals living in the United States and Canada. Four waves of data were collected, with each wave occurring approximately 1 year (M = 370 days, SD = 13) apart. Data from the first wave were not included in the current analysis because IOS data were only collected during the final three waves. Participants were initially recruited into the study using three recruitment strategies. They were active strategies (e.g., emails and listserv postings that reached participants directly through their personal email accounts), passive strategies (e.g., posting an announcement on discussion forums or classified websites), and snowball strategies (i.e., participants were provided with a link to the study that they were encouraged to share with their friends, family, and coworkers). Announcements were not placed on listservs or discussion forums specifically focused on dating, relationships, sex, or relationship problems to avoid biasing the sample. During the first wave of data collection, participants provided their email addresses and unique passwords that were used to link participants’ data from wave to wave. At subsequent waves, participants received an email directly from the first author announcing the new study wave, reminding them of their username and password, and providing a link to the study’s website. Incentive for participation at each wave was provided in the form of a lottery drawing for 1 of 20 US$100 gift cards to a popular online retailer.

Data for the current study were obtained from the subsample of individual participants who were in relationships at the second wave of the study (N = 1,659) and their follow-up data from Waves 3 and 4. Waves 2, 3, and 4 are hereafter referred to as Times 1, 2, and 3 to simplify the presentation of findings within the context of the current investigation. The sample was diverse in terms of age (M = 34.11, SD = 10), race/ethnicity (23% non-Caucasian ethnic/racial minority), sexual orientation (14% nonheterosexual), and educational attainment (54% having a 4-year college degree or greater). The sample was primarily female (78% female). Participants reported being in a diverse array of relationships with regard to relationship length (M = 8.74 years, SD = 8.27), marital status (58% married or domestic partners), and cohabitation (77% lived with their partners).

Forty-four percent (n = 732) of the originally coupled participants were retained for all waves of the study. Participants were invited to complete all waves of the survey regardless of whether they were in the same relationship, had broken up with their previous partner, and/or started a new relationship. There were no significant differences in the demographic and relationship characteristics reported above between retained participants and those who were lost to follow-up, with the exception of sexual orientation. More attrition was observed among heterosexual participants (64%) than among gay men and lesbians (50%) and bisexuals (56%), χ²(2) = 16.08, p < .001. There were also no differences between retained participants and those lost to follow-up on any of the relational well-being outcomes in the study. However, those lost to follow-up were significantly more depressed as measured by the Center for Epidemiological Studies—Depression scale (CESD; M = 9.06, SD = 6.09) than retained participants (M = 7.91, SD = 5.76), t = 4.03, p < .001.

Procedures and Measures

After consenting to participate and entering their email addresses and passwords, participants were granted access to the online questionnaire that contained the measures below.

IOS (Times 1, 2, and 3)

Aron and colleagues’ (1992) IOS scale was used to measure how individuals conceptualized their own experiences of relational closeness. This pictorial scale depicted six sets of two circles in which one circle represented the participant’s “self” and the other represented the participant’s “partner.” The sets were presented with varying degrees of overlap ranging from completely separate to almost completely overlapping. The IOS has empirically demonstrated validity and reliability (see Aron et al., 1992), most notably in its robust correlations with several other multi-item scales that measure closeness within relationships, such as the Subjective Closeness Index and the Relationship Closeness Inventory. The current study used a 2-item approach to assessing IOS, where one version of the scale assessed participants’ actual (i.e., “current”) levels of IOS and a second version assessed participants’ ideal levels of IOS (Aron, Melinat, Aron, Vallone, & Bator, 1997; Mashek & Sherman, 2004). IOS discrepancy scores were computed by subtracting each participant’s ideal IOS rating from their actual IOS rating. Negative numbers on this IOS discrepancy score indicated feeling “not close enough” to one’s partner, while positive numbers indicated feeling “too close” to one’s partner, and scores of 0 indicated no discrepancy between actual and ideal experiences of IOS.

Relationship Satisfaction (Times 1, 2, and 3)

The 4-item version of the Couples Satisfaction Index (CSI; Funk & Rogge, 2007) was included to assess individuals’ satisfaction with their current primary romantic relationships. The CSI was developed using item response theory and is the result of a factor analysis of items pooled from eight previously validated measures of relationship satisfaction. Example items include “How rewarding is your relationship with your partner?” and “In general, how satisfied are you with your relationship?” Participants responded to such items on a scale of “not at all” to “completely.” The CSI not only demonstrates strong validity correlations with existing measures of the construct, but it also demonstrates less noise and more power in detecting individual differences in satisfaction than existing measures (Funk & Rogge, 2007). The measure is scored on a scale of 0 to 21, with scores of 13.5 or below indicating relationship distress. In its originating study (Funk & Rogge, 2007), CSI (4-item) scale scores were internally consistent at .94 and were highly correlated with the eight previously validated measures of relationship satisfaction (rs ranged from .84 to .94). In the current study, participants’ scores on the CSI were internally consistent with Cronbach’s alpha coefficients of .91 at each of the three time points of the current study.

Commitment (Time 1)

Commitment was measured using the 9-item Relationship Commitment Scale (Lund, 1985). Commitment was included as a complement to relationship satisfaction given its ability to assess participants’ expectations about the future of their relationships in addition to their current levels of satisfaction. Example items read, “How likely is it that your relationship will be permanent?” and “How likely are you to pursue another relationship or single life in the future?” Participants responded to items on a scale of 1 = not at all to 7 = extremely. Final scale average scores were created that ranged from 1 to 7. In its originating study, scores on the measure demonstrated internal consistency at .82, and the scale demonstrated an ability to distinguish between other relational constructs focused on positive affect (e.g., love; Lund, 1985). Scores on the Relationship Commitment Scale in the current study were internally consistent at .76 at Time 1.

Break-Up Thoughts (Times 1, 2, and 3)

A single item was included in the survey to measure how often participants thought about ending their relationships with their current primary partners. The item read, “How often in the past month have you considered ending your relationship with your current partner?” Participants responded on a scale of 0 = never to 4 = a lot. Higher scores were considered indicative of more seriousness of participants’ considerations to dissolve their relationships, which have been demonstrated to be key determinants of actual relationship dissolution (e.g., Gottman, 1993).

Mental Health (Times 1, 2, and 3)

Mental health was assessed in the form of symptoms of depression using the short form CESD (Andresen, Malmgren, Carter, & Patrick, 1994). The CESD short form is a 10-item measure of depressive symptoms experienced over a 1-week period prior to the study view. The 10-item version uses a subset of the items from the original 20-item version (Radloff, 1977). Participants were asked how often during the past week they “could not get going,” “felt depressed,” “felt hopeful about the future,” and “felt people dislike you.” Participants responded on a 4-point scale ranging from 0 = rarely or none of the time (<1 day) to 3 = most or all of the time (5-7 days). Numerous studies have demonstrated the convergent validity of the CESD among both clinical and nonclinical samples in the form of large correlations with clinical reports of depression, Diagnostic and Statistical Manual of Mental Disorders depression diagnoses, the Hamilton Rating Scale for Depression, and the Symptom Checklist–90 (for a review of validity evidence, see McDowell & Newell, 1996; Roberts & Vernon, 1983). Although the scale has been shown to correlate moderately to highly with other measures of anxiety and psychological distress, it has been successful in identifying depression in several clinical and community samples (McDowell & Newell, 1996) and as a result is one of the most widely used measures of depressive symptoms. Internal consistency for scores on the CESD in the current study ranged from .85 to .87 over the three time points.

Analysis Plan

Cross-Sectional Segmented Regression Models

To estimate the effect of actual–ideal IOS discrepancies on relational well-being and mental health, it was necessary to fit a spline to the data. A spline, in its simplest form, can be tested in the form of a segmented regression model in which linear change is expected to be continuous up to a certain value (i.e., a knot), after which the size and direction of change will be altered. In the current study, it was hypothesized that an actual–ideal IOS discrepancy of 0—where actual closeness and ideal closeness are equal—would represent the point of expected linear change. Specifically, it was hypothesized that a positive relationship would exist between closeness discrepancies and positively valenced study outcomes (i.e., relationship satisfaction and commitment) as negative values of closeness discrepancies (i.e., feeling not close enough) approached 0. Alternatively, a negative relationship between closeness discrepancies and positively valenced outcomes was expected as positive closeness discrepancies (i.e., feeling too close) increased beyond 0. This pattern was expected to reverse for negatively valenced outcomes (i.e., break-up thoughts and depression). Put more simply, having no actual–ideal IOS discrepancy was hypothesized to correspond to optimal levels of relational well-being and mental health, with relational well-being and mental health worsening as closeness discrepancies increased beyond 0 in either direction.

Segmented linear regression models were fit to test these hypotheses (following Keele, 2008). These models included a parameter to estimate the size and direction of the relationship between closeness discrepancies and study outcomes for values of actual–ideal IOS discrepancies ranging from −6 to 0, and a separate parameter to estimate the size and direction of the relationship between actual–ideal IOS discrepancies and study outcomes for values of closeness discrepancies ranging beyond 0 to 6. This segmented linear regression model is represented in the equation below where x = an individual’s closeness discrepancy score and k = the knot discrepancy value of 0 at which the direction of linear change is expected to be altered:

Y = b_{0} + b_{1} x + b_{2} {(x)}_{+} + e

where x_{+} = {\begin{matrix} x if x > k \\ 0 if x \leq k \end{matrix}}

In this equation, b₀ is the initial intercept for an outcome Y. The sign and size of b₁ represent the direction and slope of the segment of the equation linking negative actual–ideal IOS discrepancies (i.e., feeling not close enough) to study outcomes. The sign and size of b₂ represent the direction and slope of the segment of the equation linking positive actual–ideal IOS discrepancies (i.e., feeling too close) to study outcomes. To examine the extent to which the association between actual–ideal IOS discrepancies and study outcomes remained robust and statistically significant above and beyond actual levels of closeness, actual IOS was controlled for in all segmented regression models along with gender, marital status, and length of relationship. It was not possible to include ideal IOS in these analyses because its inclusion would produce a linear dependency within the models (i.e., participants’ ideal IOS is reflected in the difference between their actual IOS and actual–ideal IOS discrepancies).

Repeated-Measures Mixed General Linear Models

To test the study’s longitudinal hypotheses, change scores were computed reflecting the change in participants’ actual–ideal IOS discrepancies over Times 1, 2, and 3 of the current study. Participants were classified into three groups for comparison based on whether the absolute value of their actual–ideal IOS discrepancies (a) diminished over time, (b) increased over time, or (c) remained the same over time. This between-subjects factor was examined within repeated-measures mixed general linear models, which examined whether change in the relational well-being and mental health outcomes over time was dependent on the type of change in participants’ actual–ideal IOS discrepancies. To examine the extent to which changes in actual–ideal IOS discrepancies corresponded to changes in study outcomes above and beyond actual levels of IOS, average levels of actual IOS across the three time points were included as a covariate given covariates in repeated-measured mixed general linear models are required to be time invariant. These models were also controlled for gender, and Time 1 marital status and length of relationship. Only participants who completed all waves of the study and reported being with the same relationship partner at each time point were included in longitudinal analyses. Missing data on variables within follow-up assessments among retained participants were minimal, with no more than 17 participants missing data on a given variable. Missing values were imputed using Expectation Maximization, an imputation technique based in Maximum Likelihood Estimation (Roth, 1994; Schafer & Olsen, 1998).

Binary Logistic Regression Models

Finally, logistic regression models were computed to examine whether actual–ideal IOS discrepancies were predictive of relationship dissolution over a 1-year period. Binary indicator variables were computed at Time 1 for those who reported higher actual IOS than ideal IOS with their partner (i.e., “too close”: 1 = yes; 0 = no) and lower actual IOS than ideal IOS with their partner (i.e., “not close enough”: 1 = yes; 0 = no). Those who reported no actual–ideal IOS discrepancies with their partners were the reference group. These binary predictor variables were entered into logistic regression models to obtain the odds ratios (ORs) regarding relationship dissolution observed over a period of 1 year associated with each type of actual–ideal IOS discrepancy. These analyses controlled for Time 1 actual IOS, gender, marital status, and length of relationship.

Results

Preliminary and Descriptive Analyses

Means, standard deviations, and bivariate Pearson correlations are presented in Table 1 for IOS-derived variables and relational well-being and mental health outcomes across the three time points in the study. As indicated by the mean levels of actual and ideal IOS, participants’ actual experiences of closeness were on average less than their ideal levels of closeness, resulting in negative mean actual–ideal IOS discrepancies at each time point. Types of actual–ideal IOS discrepancies at initial participation were not evenly distributed in the sample: 57.1% of the sample reported negative discrepancies (i.e., actual IOS < ideal IOS); 37.6% reported no discrepancies (i.e., actual IOS = ideal IOS); and 5.3% reported positive discrepancies (i.e., actual IOS > ideal IOS).

Table 1.

Descriptive Statistics and Correlations Between IOS Predictors and Relational Well-Being and Mental Health Outcomes.

	Bivariate correlations
	Actual IOS			Ideal IOS			Actual–ideal IOS discrepancy
	Time 1	Time 2	Time 3	Time 1	Time 2	Time 3	Time 1	Time 2	Time 3	M	SD
Relationship satisfaction
Time 1	.68***	.49***	.45***	.15***	.13***	.05	.61***	.41***	.45***	17.89	4.89
Time 2	.46***	.71***	.49***	.06	.19***	.04	.44***	.60***	.50***	18.11	4.71
Time 3	.43***	.51***	.69***	.08	.10*	.10*	.39***	.45***	.66***	18.31	4.57
Commitment
Time 1	.44***	.27***	.28***	.17***	.12***	.06	.36***	.19***	.26***	6.00	0.92
Break-up thoughts
Time 1	−.57***	−.37***	−.33***	−.12***	−.10**	−.07	−.52***	−.31***	−.31***	1.75	1.01
Time 2	−.34***	−.59***	−.37***	−.08*	−.15***	−.03	−.30***	−.51***	−.37***	1.70	0.97
Time 3	−.30***	−.43***	−.56***	−.05	−.11***	−.10*	−.27***	−.35***	−.53***	1.61	0.94
Depression
Time 1	−.32***	−.24***	−.24***	−.08**	−.08*	−.05	−.28***	−.20***	−.22***	8.62	5.99
Time 2	−.22***	−.37***	−.29***	−.02	−.04	−.05	−.22***	−.35***	−.27***	8.48	6.13
Time 3	−.20***	−.30***	−.34**	−.05	−.11**	−.10*	−.17***	−.23***	−.30***	8.03	6.19
M	4.37	4.41	4.43	5.32	5.24	5.27	−.96	−.84	−.83
SD	1.34	1.30	1.28	0.87	0.88	0.83	1.29	1.26	1.21

Note: IOS = inclusion of other in self.

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

Bivariate correlations demonstrated robust and statistically significant associations between actual IOS and all study outcomes, such that greater closeness was associated with higher levels of relationship satisfaction and commitment, as well as fewer break-up thoughts and depressive symptoms. Associations between ideal IOS and relational well-being and mental health outcomes—although statistically significant at times—were not robust in any instance given their corresponding effect sizes were uniformly small (i.e., Pearson rs ~ .10; Cohen, 1992). Actual–ideal IOS discrepancies demonstrated robust and statistically significant bivariate associations with relational well-being and mental health. However, these bivariate relationships are not directly interpretable given both the scaling of the discrepancy score and the unequal representation of the types of actual–ideal IOS discrepancies in the dataset, and are thus better understood via the results of the segmented regression analyses reported below.

Associations Between IOS Discrepancies and Relational Well-Being and Mental Health Outcomes

Complete results of the segmented linear regression models examining associations between actual–ideal IOS discrepancies and relational well-being and mental health are presented in Table 2. In summary, these analyses demonstrated that participants’ actual ratings of IOS were robust and statistically significant predictors of relationship satisfaction and commitment in a positive direction and break-up thoughts and depression in a negative direction. However, coefficients corresponding to the hypothesized segmented association between IOS discrepancy scores and relational well-being and mental health outcomes were robust and statistically significant above and beyond the effects of actual IOS. These segmented equations indicated that optimal levels of relational well-being (i.e., highest levels of relationship satisfaction, commitment, and lowest levels of break-up thoughts) and mental health (i.e., lowest levels of depression) occurred among those who experienced no discrepancy between their actual and ideal levels of IOS with their partners (IOS discrepancy = 0). As participants reported feeling increasingly “too close for comfort” (IOS discrepancy > 0), their reported relational well-being and mental health decreased in kind. Similarly, feeling increasingly “not close enough” (IOS discrepancy < 0) was associated with corresponding decreases in relational well-being and mental health. These associations remained robust and statistically significant controlling for gender, relationship length, and marital status. Regression equations constructed from the values presented in Table 2 were used to compute adjusted predicted scores for each outcome and visually plot the resulting segmented linear relationships, as presented in Figure 1.

Table 2.

Segmented Associations Between IOS Discrepancies and Relational Well-Being and Mental Health Outcomes.

	Relationship satisfaction					Commitment					Break-up thoughts					Depression
	B	SE	β	p	r_s	B	SE	β	p	r_s	B	SE	β	p	r_s	B	SE	β	p	r_s
Intercept	9.44	0.60		.00		4.96	0.15		.00		2.45	0.14		.00		9.85	1.01		.00
Female	0.61	0.20	.05	.00	.05	0.28	0.05	.13	.00	.12	−0.07	0.05	−.03	.17	−.03	0.24	0.34	.02	.47	.02
Relationship length	−0.04	0.01	−.06	.00	−.05	0.01	0.00	.10	.00	.09	0.00	0.00	−.04	.10	−.03	0.01	0.02	.02	.55	.01
Married	−0.02	0.20	.00	.92	.00	0.47	0.05	.25	.00	.21	−0.13	0.05	−.06	.01	−.05	−0.17	0.34	−.01	.62	−.01
Actual IOS	1.42	0.11	.39	.00	.22	0.16	0.03	.23	.00	.13	−0.21	0.03	−.27	.00	−.16	−0.65	0.18	−.15	.00	−.08
Negative IOS discrepancy (≤0)	1.65	0.12	.43	.00	.23	0.21	0.03	.30	.00	.16	−0.33	0.03	−.43	.00	−.22	−1.21	0.21	−.26	.00	−.14
Positive IOS discrepancy (>0)	−3.94	0.28	−.29	.00	−.24	−0.61	0.07	−.23	.00	−.20	0.75	0.07	.27	.00	.22	3.61	0.47	.22	.00	.18
F	321.27					114.54					178.83					42.45
R ²	.54					.35					.40					.14

Note: IOS = inclusion of other in self.

Figure 1.

Actual–ideal IOS discrepancies as a predictor of relational well-being and mental health.

Type of Change in IOS Discrepancies as a Predictor of Change in Relational Well-Being and Mental Health Over Time

Participant representation across the three levels of the between-subjects factor of type of IOS discrepancy change were as follows: 125 participants’ actual–ideal IOS discrepancies increased over time (i.e., demonstrated a net increase in a positive or negative direction, moving away from 0); 126 participants’ actual–ideal IOS discrepancies diminished over time (i.e., demonstrated a net decrease in a positive or negative direction, moving toward 0); and 260 participants’ actual–ideal IOS discrepancies remained unchanged over the 3 yearly assessments (i.e., were the same at Times 1 and 3). Means for each of these three groups defined by type of IOS discrepancy change on relational well-being and mental health outcomes are displayed in Figure 2 for each of the 3 yearly assessments.

Figure 2.

Change in relational well-being and mental health as a function of type of change in actual–ideal IOS discrepancies.

Results of the repeated-measures mixed general linear models examining the degree to which the type of change in participants’ actual–ideal IOS discrepancies was associated with changes in relational well-being and mental health over time are presented in Table 3. Only relationship satisfaction demonstrated statistically significant change over time. However, despite the lack of change in average levels of break-up thoughts and depression over time, all analyses indicated uniform and statistically significant interactions between the within-subjects factor of time and the between-subjects factor of type of change in actual–ideal discrepancies. In other words, the direction of change over time in relational well-being and mental health depended on the type of change participants experienced within their actual–ideal IOS discrepancies. These interactions (see means plots in Figure 2) indicated that, as hypothesized, those individuals whose actual experiences of IOS became closer to their ideal IOS over time demonstrated increases in relational well-being (i.e., increased satisfaction and decreased break-up thoughts) and mental health (i.e., fewer depressive symptoms) across the three time points. Those whose actual experiences of IOS moved further away from their ideal over time demonstrated decreases in relational well-being (i.e., decreased satisfaction and increased break-up thoughts) and mental health (i.e., increased depressive symptoms) across the three time points. Finally, those whose actual–ideal IOS discrepancies evidenced no change over time demonstrated relatively unchanging levels of relational well-being and mental health, remaining virtually constant over the three yearly assessments.

Table 3.

Change in Relational Well-Being and Mental Health Over Time As a Function of Type of Change in IOS Discrepancies.

	Relationship satisfaction			Break-up thoughts^a			Depression^a
Within-subjects effects	F	df	η_p²	F	df	η_p²	F	df	η_p²
Time	5.16**	2	0.01	1.34	1.97	0.00	0.19	2.00	0.00
Time × Change in actual–ideal IOS discrepancy	19.55***	4	0.07	13.86***	3.93	0.05	3.41**	4.00	0.01
Time × Actual IOS	4.52**	2	0.01	1.10	1.97	0.00	0.80	2.00	0.00
Time × Female	1.61	2	0.00	0.55	1.97	0.00	1.83	2.00	0.00
Time × Relationship length	0.27	2	0.00	1.34	1.97	0.00	0.86	2.00	0.00
Time × Married	0.68	2	0.00	2.06	1.97	0.00	1.65	2.00	0.00

Note: IOS = inclusion of other in self.

Sphericity not assumed based on results of Mauchly’s Test of Sphericity, df adjusted using Huynh–Feldt correction.

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

The time by type of change in actual–ideal IOS discrepancy interactions remained robust and statistically significant even when the effects of average actual IOS were accounted for in the models. However, only changes in relationship satisfaction over time evidenced a dependency on average levels of actual IOS. The effects of gender, relationship length, and marital status were also accounted for, although they evidenced no influence on the rate of change in relational well-being or mental health over time.

Actual–Ideal IOS Discrepancies as Predictors of Relationship Dissolution

Results of the binary logistic regression models testing the extent to which IOS discrepancies at Time 1 were predictive of relationship dissolution at follow-up assessment are presented in Table 4. Of the original Time 1 sample retained at the first yearly follow-up, 7.2% (n = 139) of participants reported ending their relationships with their previous partners. Individuals who felt “not close enough” to their partners (i.e., had negative actual–ideal IOS discrepancies) at Time 1 were approximately twice as likely to break up with their partners at follow-up than individuals who experienced no discrepancy in actual and ideal IOS at Time 1. Similarly, individuals who felt “too close” to their partners (i.e., had positive actual–ideal IOS discrepancies) at Time 1 were approximately twice as likely to break up with their partners over 1 year than individuals who experienced no discrepancy in actual and ideal IOS at Time 1; however, this association was not statistically significant. These findings were obtained from models controlling for actual IOS at Time 1, which demonstrated a protective effect on relationship dissolution along with relationship length and marital status.

Table 4.

Actual–Ideal IOS Discrepancies as Predictors of Relationship Dissolution.

	B	p	OR	95% CI for OR
Intercept	0.24	.71	1.28
Negative IOS discrepancy	0.73	.04	2.07	[1.04, 4.14]
Positive IOS discrepancy	0.77	.15	2.16	[0.75, 6.18]
Actual IOS	−0.33	.00	0.72	[0.58, 0.88]
Female	−0.12	.66	0.89	[0.52, 1.52]
Relationship length	−0.21	.00	0.81	[0.74, 0.88]
Married	−0.99	.00	0.37	[0.19, 0.72]

Note: IOS = inclusion of other in self; OR = odds ratio; CI = confidence interval.

Discussion

The findings of the current study demonstrate the robust utility of the construct of IOS in predicting relational well-being, stability, and mental health among a diverse sample of individuals in romantic relationships. Consistent with previous theory and research (Aron et al., 2004; Tsapelas et al., 2009), experiences of closeness with one’s relationship partner—as evidenced via greater inclusion of one’s partner within one’s own self-concept—is indicative of heightened relational well-being across a variety of indicators, including increased satisfaction, commitment, and fewer thoughts about ending one’s relationship. However, this study’s findings indicate that individuals’ actual experiences of IOS in their relationships are only part of a broader experience of closeness, which must be accounted for in attempts to fully understand the association between closeness and well-being in romantic relationships. Specifically, these findings indicate that optimal levels of relational well-being and mental health are achieved when individuals’ actual experiences of closeness in their relationships match their idealized levels of closeness with their relationship partners. This finding provides further evidence for existing theory and research on the phenomena of feeling “too close” in romantic relationships (Mashek & Sherman, 2004) by providing an overarching explanatory model of closeness discrepancies and their impact on individual health and romantic relationship quality.

The current study extends this line of research by demonstrating that individuals who feel “too close” and “not close enough” to their partner—as operationalized via actual–ideal IOS discrepancies—fare increasingly less well as the size of their IOS discrepancies increase, regardless of what their actual experiences of IOS may be. This finding was consistent not only across a variety of indicators of relational well-being, but extended to the domain of mental health as well. Thus, accounting for individuals’ actual experiences of closeness with their partners in relation to their desired levels of closeness is necessary to maximize the success of attempts to understand the effect of closeness on relational well-being and mental health in relationships. What seems to matter most for both relational well-being and mental health is not only how close people actually feel to their partners, but whether people’s actual experiences of closeness in their relationships match how close they would like to feel to their partners.

Furthermore, this study provides empirical evidence that the abatement of discrepancies between experienced and idealized levels of closeness in relationships can lead to improved relational well-being, stability, and mental health over time. Specifically, the findings from the current study show that if experiences of closeness with a partner move in line with ideal experiences of closeness, individuals will experience significant and substantial increases in relational well-being and mental health over time. However, the opposite is true under conditions where experiences of closeness move further away from ideal levels of closeness over time, resulting in decreased relational well-being and mental health. Those whose discrepancies remained constant over time did not experience much change in their relational well-being or mental health over time. It is important to note that the effects of participants’ closeness discrepancies on changes in relational well-being and mental health persisted regardless of how close they actually felt to their partner. This novel finding demonstrates the impact and importance of closeness discrepancy reduction—above and beyond the promotion of closeness itself—in efforts to improve relational well-being and mental health over time.

Even further, experiencing a discrepancy in actual–ideal levels of IOS was associated with an increased likelihood of relationship dissolution, above and beyond actual experiences of closeness. Even though feeling “not close enough” to a partner was significantly associated with dissolution rates and feeling “too close” to a partner was not significantly associated with dissolution rates, it is important to note that the effect sizes for both were nearly equal (ORs = 2.07 and 2.16, respectively). Thus, it is not possible to conclude that there were differential effects of the type of closeness discrepancy on relationship dissolution. The magnitude and direction of the effects associated with both types of closeness discrepancies indicate that individuals with any type of discrepancy between their actual and idealized experiences of closeness are at substantial risk for relationship dissolution compared with individuals who experience no discrepancy.

Accounting for the existence and size of the discrepancy between individuals’ actual and ideal levels of closeness in their relationships is paramount in attempts to explain its effects on relational well-being and mental health. Less important in this regard is the nature or direction of the discrepancy (Aron et al., 2004; Mashek & Sherman, 2004). In other words, optimal levels of relational well-being and mental health exist when actual–ideal closeness discrepancies do not. Feeling both “too close” and feeling “not close enough” to a relationship partner have detrimental effects on relational well-being and mental health. The deleterious implications posed by the encroachment on personal identity and control likely associated with feeling “too close” to one’s partner seem to be equal to those posed by the lack of closeness experienced in the form of less-than-desirable levels of closeness.

Understanding individuals’ experiences of closeness in their romantic relationships via their reports of actual–ideal IOS discrepancies has tremendous potential not only for future research on closeness, IOS, and health and well-being in close relationships, but for clinical and counseling work with couples and individuals as well. The present findings highlight necessary next sttif in research on IOS and relational well-being and mental health, which have thus far overly relied on single indicators of IOS in relational contexts. Measuring actual–ideal IOS discrepancies takes seriously individuals’ own desires for closeness in relationships and avoids forcing couples’ lived experiences into a researcher-imposed expectation of what their relationships “should” look like (e.g., healthier relationships = more closeness). It is this kind of constructivist approach to understanding closeness in relationships that, as the current study demonstrates, provides an improved explanation of relational functioning over what static measures of closeness can explain by themselves.

Accounting for actual–ideal IOS discrepancies in the clinical intake process and assessment of couples counseling interventions may thus prove to be a useful tool for clinicians seeking to improve relational well-being and mental health in relational contexts. Having couples reflect on their current and ideal levels of closeness on multiple occasions can assist clinicians and counselors in their abilities to target interventions aimed at eliminating closeness discrepancies and problematic experiences of enmeshment and intrusiveness, thereby improving the health and well-being of their patients and clients (e.g., Green & Werner, 1996). Furthermore, the present findings suggest that efforts focused on improving relational well-being and mental health in couples should not limit themselves to the promotion of closeness but instead focus on the alignment of individuals’ experiences of closeness and their ideal levels of closeness with their partners.

Limitations and Suggestions for Future Research

The findings of the current study should be interpreted in light of some limitations. Although the study reports on longitudinal findings, the data are nonetheless correlational in nature and causality cannot be inferred. Second, the sample is biased by self-selection and is not random or representative of the U.S. and Canadian populations from which it was drawn. Although research indicates men and women do not differ in their experiences of IOS discrepancies (Frost & Eliason, in press) and the effects of gender were controlled for in all analyses, the underrepresentation of men in the study did not permit an examination of potential gender differences in the associations between IOS discrepancies and study outcomes. Despite these limitations, it is important to note that the size and diversity of the present sample provides a complement to the existing research on closeness and IOS, which has been conducted primarily with undergraduate student samples or geographically restricted samples of mostly married individuals.

Significant amounts of attrition were observed that were likely due to the challenges of conducting longitudinal research on the Internet and the minimal incentive structure used. However, there were very few differences between retained participants and those lost to follow-up. As a result of higher attrition among those who were more depressed, the effect sizes of coefficients linking IOS discrepancies with mental health are likely attenuated and thus underestimated in the current study. Also, associations between IOS discrepancies and dissolution are likely attenuated, given individuals who break up with their partners may be less motivated to continue participation in a study about relationships than those who stay together. In addition, the lack of statistical significance associated with the robust OR linking positive actual–ideal IOS discrepancies to dissolution was likely a result of the low frequency of positive IOS discrepancies in the sample, as well as the difficulty in observing relationship dissolution in survey studies with yearly assessments (Gottman, 1993).

The study used a limited set of indicators of relational well-being and mental health; thus, future work is needed to examine the extent to which the current findings generalize to other indicators of relationship quality and health. Given depressive symptoms are just one indicator of a broader construct of mental health, future studies should include additional measures of negative (e.g., anxiety) and positive functionality (e.g., psychological well-being). Also, the current study examined only individuals in romantic relationships. Future studies should explore how processes related to closeness and IOS discrepancies in relationships play out within couples. For example, the use of dyadic analyses will allow for the examination of research questions surrounding partner effects of IOS discrepancies on relational well-being and mental health (e.g., How do husbands’ IOS discrepancies impact their wives’ depression?). Additional research is also needed to explore the implications of actual–ideal IOS discrepancies in other types of close relationships given the centrality of closeness to positive relations with others. Important research questions can be posed regarding the impact of actual–ideal IOS discrepancies on a multitude of out- comes across a variety of relationship contexts such as job performance among coworkers, familial conflict in parent–child relationships, and satisfaction with clinical care in patient–provider interactions.

Conclusion

The present study’s focus on actual–ideal closeness discrepancies deepens psychological understandings of the effects of closeness on relationship quality and individual health in the context of romantic relationships. Accounting for individuals’ actual experiences of closeness in romantic relationships is necessary but not sufficient in effectively explaining their relational well-being, stability, and mental health. Psychological inquiries into the phenomenon of closeness must additionally account for individuals’ idealized notions of closeness in their relationships with their partners, given discrepancies between actual and ideal experiences of closeness can be detrimental to relational well-being and mental health, both in the present and over time. Future work must take seriously where individuals’ experiences of closeness with their partners are in relation to where they would ideally like them to be to more fully understand and promote their experiences of closeness, well-being, health, and relational stability over time.

Footnotes

Acknowledgements

The authors thank members of the Psychology of Sexuality and Relationships Lab and the Columbia and NYU Couples Labs for their comments on presentations of the data reported in this article.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Agnew

C. R.

Loving

T. J.

Goodfriend

(2004). Thinking close: Measuring closeness as perceived self-other inclusion. In Mashek

D. J.

Aron

(Eds.), Handbook of closeness and intimacy (pp. 103-116). Mahwah, NJ: Lawrence Erlbaum.

Agnew

C. R.

Van Lange

P. A. M.

Rusbult

C. E.

Langston

C. A.

(1998). Cognitive interdependence: Commitment and the mental representation of close relationships. Journal of Personality and Social Psychology, 74, 939-954.doi:10.1037/0022-3514.74.4.939

Andresen

E. M.

Malmgren

J. A.

Carter

W. B.

Patrick

D. L.

(1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D. American Journal of Preventive Medicine, 10, 77-84.

Aron

E. N.

Smollan

(1992). Inclusion of Other in the Self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63, 596-612.

Aron

E. N.

Tudor

Nelson

(1991). Close relationships as including other in the self. Journal of Personality and Social Psychology, 60, 241-253. doi:10.1037//0022-3514.60.2.241

Aron

McLaughlin-Volpe

Mashek

Lewandowski

Wright

S. C.

Aron

E. N.

(2004). Including others in the self. European Review of Social Psychology, 15, 101-132. doi:10.1080/10463280440000008

Aron

Melinat

Aron

Vallone

R. E.

Bator

R. J.

(1997). The experimental generation of interpersonal closeness: A procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23, 365-378.

Aron

Norman

C. C.

Aron

E. N.

(2001). Shared self-expanding activities as a means of maintaining and enhancing close romantic relationships. In Harvey

Wenzel

(Eds.), Close romantic relationships: Maintenance and enhancement (pp. 47-66). Mahwah, NJ: Lawrence Erlbaum.

Bartholomew

Horowitz

L. M.

(1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244.

10.

Berscheid

(2010). Love in the fourth dimension. Annual Review of Psychology, 61, 1-25.

11.

Birnbaum

Reis

Mikulincer

Gillath

Orpaz

(2006). When sex is more than just sex: Attachment orientations, sexual experience, and relationship quality. Journal of Personality and Social Psychology, 91, 929-943.

12.

Brennan

K. A.

Shaver

P. R.

(1995). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21, 267-283.

13.

Brunell

A. B.

Pilkington

C. J.

Webster

G. D.

(2007). Perceptions of risk in intimacy in dating couples: Conversation and relationship quality. Journal of Social & Clinical Psychology, 26, 92-119.

14.

Butzer

Campbell

(2008). Adult attachment, sexual satisfaction, and relationship satisfaction: A study of married couples. Personal Relationships, 15, 141-154.

15.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155-159.

16.

Fitzsimons

G. M.

Kay

A. C.

(2004). Language and interpersonal cognition: Causal effects of variations in pronoun usage on perceptions of closeness. Personality and Social Psychology Bulletin, 30, 547-557. doi:10.1177/0146167203262852

17.

Fletcher

G. J. O.

Simpson

J. A.

(2000). Ideal standards in close relationships: Their structure and functions. Current Directions in Psychological Science, 9, 102-105.

18.

Fletcher

G. J. O.

Simpson

J. A.

Thomas

Giles

(1999). Ideals in intimate relationships. Journal of Personality and Social Psychology, 76, 72-89. doi:10.1037/0022-3514.76.1.72

19.

Frost

D. M.

Eliason

(in press). Challenging the assumption of fusion in female same-sex relationships. Psychology of Women Quarterly.

20.

Frost

D. M.

(in press). The narrative construction of intimacy and affect in relationship stories: Implications for relationship quality and mental health. Journal of Social and Personal Relationships. DOI: 10.1177/0265407512454463

21.

Funk

J. L.

Rogge

R. D.

(2007). Testing the ruler with item response theory: Increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21, 572-583.

22.

Goodboy

A. K.

Booth-Butterfield

(2009). Love styles and desire for less closeness in romantic relationships. Psychological Reports, 105, 191-197.

23.

Gottman

J. M.

(1993). A theory of marital dissolution and stability. Journal of Family Psychology, 7, 57-75.

24.

Green

Werner

P. D.

(1996). Intrusiveness and closeness-caregiving: Rethinking the concept of family “Enmeshment.” Family Process, 35, 115-136. doi:10.1111/j.1545-5300.1996.00115.x

25.

Hassebrauck

Fehr

(2002). Dimensions of relationship quality. Personal Relationships, 9, 253-270. doi:10.1111/1475-6811.00017

26.

Hazan

Shaver

P. R.

(1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry, 5, 1-22.

27.

Heidrich

S. M.

Powwattana

(2004). Self-discrepancy and mental health in older women with chronic illnesses. Journal of Adult Development, 11, 251-259.

28.

Higgins

E. T.

(1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340.

29.

Holt-Lunstad

Smith

T. B.

Layton

J. B.

(2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7, e1000316. doi:10.1371/journal.pmed.1000316

30.

Kashdan

T. B.

Volkmann

J. R.

Breen

W. E.

Han

(2007). Social anxiety and romantic relationships: The costs and benefits of negative emotion expression are context dependent. Journal of Anxiety Disorders, 21, 475-492.

31.

Keele

L. J.

(2008). Semiparametric regression for the social sciences. New York, NY: John Wiley.

32.

Khaleque

(2004). Intimate adult relationships, quality of life and psychological adjustment. Social Indicators Research, 69, 351-360. doi:10.1007/s11205-004-1543-x

33.

Kiecolt-Glaser

J. K.

Gouin

J.-P.

Hantsoo

(2010). Close relationships, inflammation, and health. Neuroscience & Biobehavioral Reviews, 35, 33-38.doi:10.1016/j.neubiorev.2009.09.003

34.

Dove

N. L.

Agnew

C. R.

Korn

M. S.

Mutso

A. A.

(2010). Predicting nonmarital romantic relationship dissolution: A meta-analytic synthesis. Personal Relationships, 17, 377-390. doi:10.1111/j.1475-6811.2010.01285.x

35.

Lewandowski

G. W.

Nardone

Raines

A. J.

(2010). The role of self-concept clarity in relationship quality. Self and Identity, 9, 416-433. doi:10.1080/15298860903332191

36.

Lund

(1985). The development of investment and commitment scales for predicting continuity of personal relationships. Journal of Social and Personal Relationships, 2, 3-23. doi:10.1177/0265407585021001

37.

Mashek

D. J.

Israel

Aron

(2011). Wanting less closeness in romantic relationships. Basic and Applied Social Psychology, 33, 333-345.

38.

Mashek

D. J.

Sherman

M. D.

(2004). Desiring less closeness with intimate others. In Mashek

D. J.

Aron

(Eds.), Handbook of closeness and intimacy (pp. 343-356). Mahwah, NJ: Lawrence Erlbaum

39.

McDowell

Newell

(1996) Measuring health: A guide to rating scales and questionnaires (2nd ed.). New York, NY: Oxford University Press.

40.

Patrick

Knee

C. R.

Canevello

Lonsbary

(2007). The role of need fulfillment in relationship functioning and well-being: A self-determination theory perspective. Journal of Personality and Social Psychology, 92, 434-457.

41.

Radloff

L. S.

(1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

42.

Reis

H. T.

Aron

(2008). Love: What is it, why does it matter, and how does it operate? Perspectives on Psychological Science, 3, 80-86.

43.

Reis

H. T.

Collins

W. A.

Berscheid

(2000). The relationship context of human behavior and development. Psychological Bulletin, 126, 844-872. doi:10.1037/0033-2909.126.6.844

44.

Roberts

R. E.

Vernon

S. W.

(1983). The Center for Epidemiologic Studies Depression Scale: Its use in a community sample. American Journal of Psychiatry, 140, 41-46.

45.

Robins

Boldero

(2003). Relational discrepancy theory: The implications of self-discrepancy theory for dyadic relationships and for the emergence of social structure. Personality and Social Psychology Review, 7, 56-74.

46.

Robles

T. F.

Kiecolt-Glaser

J. K.

(2003). The physiology of marriage: Pathways to health. Physiology & Behavior, 79, 409-416. doi:10.1016/S0031-9384(03)00160-4

47.

Roth

P. L.

(1994). Missing data: A conceptual review for applied psychologists. Personnel Psychology, 47, 537-560.

48.

Rusbult

C. E.

Van Lange

P. A. M.

(2003). Interdependence, interaction, and relationships. Annual Review of Psychology, 54, 351-375.

49.

Ryan

Deci

(2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78.

50.

Schafer

J. L.

Olsen

M. K.

(1998). Multiple imputation for multivariate missing-data problems: A data analyst’s perspective. Multivariate Behavioral Research, 33, 545-571.

51.

Scott

O’Hara

M. W.

(1993). Self-discrepancies in clinically anxious and depressed university students. Journal of Abnormal Psychology, 102, 282-287.

52.

Seider

B. H.

Hirschberger

Nelson

K. L.

Levenson

R. W.

(2009). We can work it out: Age differences in relational pronouns, physiology, and behavior in marital conflict. Psychology and Aging, 24, 604-613. doi:10.1037/a0016950

53.

Sheldon

K. M.

Williams

Joiner

(2003). Self-determination theory in the clinic: Motivating physical and mental health. New Haven, CT: Yale University Press.

54.

Tsapelas

Aron

Orbuch

(2009). Marital boredom now predicts less satisfaction 9 years later. Psychological Science, 20, 543-545. doi:10.1111/j.1467-9280.2009.02332.x

55.

Werner

P. D.

Green

R. J.

Greenberg

Browne

T. L.

McKenna

T. E.

(2001). Beyond enmeshment: Evidence for the independence of intrusiveness and closeness-caregiving in married couples. Journal of Marital & Family Therapy, 27, 459-472.