Which of the following is identified as an element of constraint in the commitment of marriage?

J Fam Psychol. Author manuscript; available in PMC 2011 Oct 1.

Published in final edited form as:

PMCID: PMC2958669

NIHMSID: NIHMS227801

Abstract

Many have argued that it is important to examine different aspects of commitment in romantic relationships, but few studies have done so. Using a large, national sample of unmarried adults in relationships (N = 1184), this study examined four aspects of relationship commitment and their associations with relationship adjustment and stability. We examined dedication (i.e., interpersonal commitment) as well as three types of constraint commitment: perceived constraints (e.g., social pressure to stay together or difficulty of termination procedures, measured using Stanley and Markman’s (1992) Commitment Inventory), material constraints (e.g., signing a lease, owning a pet), and felt constraint (i.e., feeling trapped). Cross-sectionally, these four facets of commitment were associated in expected directions with relationship adjustment, as well as perceived likelihood of relationship termination and of marriage. Longitudinally, each facet uniquely predicted relationship stability. More dedication, more material and perceived constraints and less felt constraint were uniquely associated with a higher likelihood of staying together over an eight-month period.

Keywords: Commitment, constraint, investment, unmarried, dating, relationship adjustment, marriage

Theorists have argued that understanding commitment is fundamental to understanding romantic relationship development and relationship quality (Johnson, 1973; Rusbult, 1980; Stanley & Markman, 1992). At the same time, little research has focused on determining which specific aspects of commitment are most predictive of relationship continuance versus termination. The purpose of the present study was to explore different facets of relationship commitment in unmarried relationships and how they related concurrently to other relationship characteristics as well as to relationship stability over time. Specifically, we examined dedication (i.e., interpersonal commitment) as well as three aspects of constraint commitment (i.e., factors that can serve to keep someone in a relationship when they might rather leave).

Theories of Commitment

Most modern views of commitment find their historical roots in interdependence theory or social exchange theories. The former was developed by social psychologists (e.g., Kelley & Thibaut, 1978; Levinger, 1965; Thibaut & Kelley, 1959) and the latter by sociologists and economists (e.g., Cook & Emerson, 1978; Emerson, 1962; Homans, 1958). There is great overlap between these two theoretical systems, with interdependence theory growing out of the foundations of exchange theory. These theoretical systems are concerned with the forces that attract partners, and the personal, interpersonal, and social factors that influence the formation, development, and continuance of relationships. These theoretical perspectives have generated a number of studies on the nature and function of commitment (e.g., Kurdek, 1995) as well as more specific theories about the nature of commitment in relationships (see Adams & Jones, 1999). Of particular relevance for this paper, Levinger (1965) focused on attraction and barrier forces, grounded in an interdependence framework, and Thibaut and Kelley (1959) posited that stability of relationship was not only a function of attraction, but the comparison level for alternatives. In other words, sometimes people remain in relationships they might rather leave because the barriers to leaving are too great or the alternatives to staying are poor. Hence, running through these conceptions was the idea that people stay or leave relationships for reasons additional to their satisfaction in the relationship and their desire to stay.

One early approach to understanding commitment that applied ideas from these broader theories was developed by Johnson (1973). This tripartite model distinguished between personal, moral, structural commitment. Personal commitment is the sense that one wants to stay in the relationship, whereas moral commitment is a sense of obligation to stay (e.g., believing that divorce is generally wrong), and structural commitment refers to constraints or pressures that could make partners remain together (Johnson, Caughlin, & Huston, 1999). Structural commitment encompasses forces such as the quality of alternatives to the present relationship and the degree of difficulty of the specific steps required to end a relationship. Research supports the distinction between personal commitment and moral or structural commitment (e.g., Adams & Jones, 1997). These aspects of commitment tend to relate to relationship maintenance behaviors in different ways (Ramirez, 2008).

Based on interdependence theory, social psychologist Caryl Rusbult developed a system for understanding commitment under the moniker of the investment model. The investment model suggests that commitment is a function of satisfaction with the relationship, the perceived quality of alternatives, and the perceived investments one has made in the relationship (Rusbult, 1980, 1983). This model has generated a great deal of research, to the point that meta-analyses now support the model’s utility. Across numerous studies and even different relationship types, these three elements are associated with commitment (Le & Agnew, 2003). Following this model, one study examined how different kinds of investments were associated with relationship termination (Goodfriend & Agnew, 2008). This study made distinctions between tangible (e.g., joint bank account) and intangible (e.g., self-disclosures) investments as well as planned and past investments. They found that measuring these investments separately was more useful in explaining variability in relationship outcomes than was a global measure of investments.

Stanley and Markman (1992) also presented a framework for understanding commitment that drew upon these various conceptual schemes and they validated assessment tools for different aspects of commitment. Consistent with the interdependence model that emphasized attraction and barrier forces (Levinger, 1965), Stanley and Markman suggested that two meta-constructs, dedication and constraint, are most important to understanding commitment. This basic emphasis has the utilitarian value of focusing on forces based on the “want to” versus the “have to” of commitment dynamics (also see Stanley, Rhoades, & Whitton, in press).

Dedication refers to forces that play a role in the motivation to build and maintain the quality of the relationship, including factors such as wanting a long-term future with the partner, having an identity as a couple, and willingness to sacrifice for the partner or relationship (Stanley & Markman, 1992). In general terms, dedication can be thought of as interpersonal commitment or the desire to maintain the relationship with the current partner. Because it is based in this desire, dedication is typically associated with higher relationship satisfaction (e.g., Owen, Rhoades, Stanley, & Markman, in press) and relationship stability (Impett, Beals, & Peplau, 2001; Le & Agnew, 2003). Dedication is similar to the general construct of commitment in the investment model (cf. Rusbult, Martz, & Agnew, 1998) or personal commitment in the tripartite model (cf. Johnson et al., 1999).

Constraints refer more to factors that reinforce staying in a relationship even if satisfaction and dedication wane, such as financial or other tangible investments one has made in the relationship, concern for the partner’s welfare, or a sense of moral obligation to stay (Stanley & Markman, 1992). Dedication and constraint commitment are moderately positively associated, both because dedication likely leads to behaviors that increase constraints and because constraints provide a psychological motivation (fear of loss) to keep engaging in dedicated behavior (cf. Simon, Krawczyk, & Holyoak, 2004). Thus, constraint commitment is not inherently bad or good. Making investments or being aware of constraints in a relationship does not lead one to feel overly constrained or trapped unless dedication and relationship quality are lacking (Stanley & Markman, 1992).

Three Aspects of Constraint Commitment

In the current study, we measure dedication as suggested by Stanley and Markman (1992), but we extend existing models of commitment by expanding the conceptualization and measurement of constraints. We examine three ways of indexing constraints and how each aspect predicts relationship stability along with dedication. Typically, investments or constraints have been measured globally (e.g., I have made many investments in this relationship) and as perceptions or appraisals (e.g., my friends would not mind if my partner and I broke up). We believe these existing assessments of constraints are useful, but that additional information regarding constraints may add to the field’s understanding of commitment and processes related to relationship stability. Thus, we also examined a more objective measure of constraints in which respondents indicate if they have made particular material investments that may make a couple more constrained, such as sharing debt. This measure of what we call material constraints taps relatively factual information and therefore does not require any kind of appraisal of the impact of the investment on the relationship. On the other end of the spectrum, we also included a measure that requires respondents to appraise the potential impact of constraints by asking participants whether they feel trapped in their relationships. We next describe these aspects of constraints in more detail.

Perceived constraint commitment refers to internal or external forces that encourage partners to stay together (Stanley & Markman, 1992) and represents the way constraint commitment has typically been measured in this field. Examples of this aspect of constraint commitment are social pressure to stay together, the sense that one has made tangible and non-tangible investments in the relationship that would be lost if the relationship ended, believing that one’s quality of life would deteriorate should the relationship end, feeling concerned for the welfare of one’s partner, thinking the steps to end the relationship would be difficult, and believing that it would be difficult to find a suitable mate if the relationship ended. We call these perceived constraints because they are recognized and assessed by partners as factors that could be relevant to relationship continuance. In the present study, we use updated versions of Stanley and Markman’s (1992) Constraint Commitment Scale items that have been recently validated (Owen et al., in press). Each item asks respondents to rate the degree to which he or she agrees or disagrees with statements such as, “I have put a number of tangible, valuable resources into this relationship” or “the process of ending this relationship would require many difficult steps.” Thus, these items ask respondents to recognize constraints they may have in their relationships, but not judge them as positive or negative.

Material constraints reflect investments that couples may or may not directly perceive as potential constraints but that nevertheless may create forces that encourage the relationship to continue. These material constraints can be measured more objectively because, in contrast to many perceived constraints, they are specific, tangible resources that a couple shares. Examples of such investments are sharing debt, signing a lease, buying furniture, listing a partner as a beneficiary, owning a pet together, or having made plans for a vacation in the future. Material constraints were measured in the current study using a checklist on which respondents indicated which items they share (e.g., “our names are listed on a lease together”). Partners in the same relationship are likely to have high concordance on measures of material constraints, as they do not require an appraisal of how the investment is related to relationship dynamics or maintenance. The concept of material constraints is similar to what Goodfriend and Agnew (2008) call tangible investments and to Stanley and Markman’s (1992) subtype of perceived constraint called structural investments, although it was measured quite differently in the current study because respondents were asked to indicate specific investments rather than providing a general appraisal of them. Further, the items used to measure material constraints were developed for couples in premarital, developing relationships (Rhoades, Stanley, & Markman, 2010).

The third aspect of constraint commitment assessed in the current study is felt constraint. Felt constraint refers to the sense that one is constrained in the relationship due to external pressures. Compared to perceived and material constraints, which can be thought of as more objectively measured forms of constraint commitment, felt constraint is a personal appraisal of how investments and barriers to leaving are affecting whether the relationship continues or not. Adams and Jones (1997) measured a construct called “feelings of entrapment” in their work, though this construct did not assess whether respondents felt trapped in the way the present study did. Instead, their feelings of entrapment measure included items that we would consider indicators of perceived constraint commitment, correlating most uniquely in their study with Stanley and Markman’s (1992) measure of structural investments. The current study’s measure of felt constraint asked respondents to directly rate how trapped or stuck they feel by the investments they have made in their relationship. Felt constraint should be negatively related to a desire to maintain the relationship (i.e., dedication) because only individuals who want out of the relationship should feel that they are negatively impacted by constraints. With regard to relationship termination, felt constraint may function differently from the other two aspects of constraint commitment in that feeling constrained may predict ending the relationship rather than maintaining it.

Present Study

This study examined the four different facets of commitment described above (i.e., dedication, perceived constraint, felt constraint, and material constraints) in a large nationwide sample of unmarried individuals aged 18 to 35. It was not our intention to build or test a new theoretical model regarding commitment, rather, our goal was to assess components of existing constructs in the commitment literature that have not received much or any prior attention. We first examined how these facets of commitment were related, cross-sectionally, to each other and to other relationship characteristics including length of relationship, relationship adjustment, as well as perceived likelihood of relationship dissolution and marriage. Next, we examined how these facets of commitment were related to relationship stability over time. Given their distinct natures, we hypothesized that each of the four major facets of commitment would uniquely predict break-up eight months following the initial assessment of commitment.

Method

Participants

Participants (N = 1184) in the current study were individuals who took part in the first three waves of a larger, longitudinal project on romantic relationship development. The current sample included 425 men (35.9%) and 759 women. At the initial wave of data collection, participants ranged in age from 18 to 35 (M = 25.60 SD = 4.80), had a median of 14 years of education and made $15,000 to $19,999 annually, on average. All participants were unmarried but in a romantic relationship with someone of the opposite sex at the initial assessment (M length of relationship = 34.63 months, SD = 34.03); 32.3% were cohabiting (defined as sharing a single address without either partner having a separate place). In terms of ethnicity, this sample was 7.9% Hispanic or Latino and 92.1% not Hispanic or Latino. In term of race, the sample was 76.3% White, 14.4% Black or African American, 3.3% Asian, 1.2% American Indian/Alaska Native, and .2% Native Hawaiian or Other Pacific Islander; 3.6% reported being of more than one race and 1.0% did not report a race. With regard to children, 33.6% of the sample reported that there was at least one child involved in their romantic relationship. Specifically, 13.7% of the sample had at least one biological child together with their current partner, 17% had at least one biological child from previous partner(s), and 19.3% reported that their partner had at least one biological child from previous partner(s).

Procedure

To recruit participants for the larger project, a calling center used a targeted-listed telephone sampling strategy to call households within the contiguous United States. After a brief introduction to the study, respondents were screened for participation. To qualify, respondents needed to be between 18 and 34 and be in an unmarried relationship with a member of the opposite sex that had lasted two months or longer. Those who qualified, agreed to participate, and provided complete mailing addresses (N = 2,213) were mailed forms within two weeks of their phone screening. Of those who were mailed forms, 1,447 individuals returned them (65.4% response rate); however, 152 of these survey respondents indicated on their forms that they did not meet requirements for participation, either because of age or relationship status, leaving a sample of 1295 for the first wave (T1) of data collection. These 1295 individuals were mailed the second wave (T2) of the survey four months after returning their T1 surveys. The third wave (T3) was mailed four months after T2.

For the current study, relationship stability data were obtained from T2 and T3. At these time points, participants were asked whether they were still together with the person they were dating the last time they completed forms for the study. Individuals who were broken up with the person they had been dating at T1 by either T2 or T3 were included in the broken up group. To be included in the intact group, individuals needed to have completed T3 and indicated then that they were in the same relationship from T1. Of the initial sample, 1184 (91.4%) participants met criteria for one of these groups. The other 9.6% were lost to attrition. Of the final 1184, 873 (73.7%) remained in the same romantic relationship over the eight-month period and 311 (26.3%) broke up within the 8-month timeframe.

Measures

Dedication

Dedication was measured with 14 items from the Commitment Inventory (Stanley & Markman, 1992). The Commitment Inventory was originally developed using novel items and a few items that were revisions of items from Johnson’s work (1973). Factor analyses and comparisons across samples supported its factor structure and validity (Stanley & Markman, 1992). Since the original publication of this inventory, Stanley has made several revisions, including the addition of new items, revisions of the response scale, and a total dedication score rather than several subscales of this construct. This new version has been shown to be reliable and valid in other research (e.g., Kline et al., 2004; Owen et al., in press). For the dedication subscale, each item was rated on a 1 (strongly disagree) to 7 (strongly agree) scale. Example items are “I want this relationship to stay strong no matter what rough times we encounter,” “I like to think of my partner and me more in terms of ‘us’ and ‘we’ than ‘me’ and ‘him/her,’” “It makes me feel good to sacrifice for my partner,” and “My relationship with my partner is clearly part of my future life plans.” A mean score was used in the analyses and higher scores are indicative of more dedication. Scores could range from 1 to 7. In this sample, the measure was internally consistent with a Cronbach’s alpha (α) of .88.

Perceived constraint

To measure perceived constraints, we used the total score from the constraint scale of the revised version of the Commitment Inventory (Owen et al., in press; Stanley & Markman, 1992), except for the items related to concern for children’s welfare because not all participants had children. The total score included 19 items. Example items are: “It would be difficult for my friends to accept it if I ended the relationship with my partner,” “The steps I would need to take to end this relationship would require a great deal of time and effort,” “I could not bear the pain it would cause my partner to leave him/her even if I really wanted to,” and “I have put a number of tangible, valuable resources into this relationship”. The response scale ranged from 1 (strongly disagree) to 7 (strongly agree). The mean score was used in analyses and higher scores reflect more perceived constraints. For this sample, α = .80.

Material constraints

To measure material constraints, we used The Joint Activities Checklist (Rhoades et al., 2010). It includes 25 external factors that may serve to reinforce individuals staying together, such as owning a house together, paying for each other’s credit cards, having a pet, having paid for future vacation plans, making home improvements together, signing a lease, or having a joint-bank account. It was designed as an objective measure of constraints and Pearson correlations demonstrated high within-couple reliability (r = .82) in previous research (Rhoades et al., 2010). Internal consistency was high in the current sample, α = .85. A sum of the items checked was used in the analyses, thus scores could range from 0 to 25.

Felt constraint

This scale included three items: “I feel trapped in this relationship but I stay because I have too much to lose if I leave,” “I would leave my partner if it was not so difficult to do,” and “I feel stuck in this relationship.” Each was measured on a 1 (strongly disagree) to 7 (strongly agree) response scale. The first item has been used in other research (Stanley, Markman, & Whitton, 2002), but the others are novel.

Internal consistency was high, α = .82.

Relationship adjustment

To measure relationship adjustment, we used the 4-item version of the Dyadic Adjustment Scale (Sabourin, Valois, & Lussier, 2005; Spanier, 1976). This measure includes items about thoughts about dissolution, frequency of confiding in one another, and a general item about the degree of happiness with the relationship. In this sample, α = .81. It was scored by summing the items and higher scores indicate higher relationship adjustment. Scores could range from 0 to 21.1

Perceived likelihood of dissolution

The relationship instability item from the National Survey of Families and Households was used to measure participants’ predictions about future relationship dissolution. The item asked respondents to assess the probability that the relationship would dissolve on a 5-point Likert scale (i.e., “What is the likelihood that you and your partner will break-up within the next year?”). Internal consistency could not be calculated for this single item measure, but test-retest reliability has been shown to be high in other research with a similar sample (Rhoades et al., 2010). Scores could range from 1 to 5. In this sample, actual scores ranged from 1 to 5 (M = 1.58, SD = .80).

Perceived likelihood of marriage

A continuous item, “How likely is it that you and your partner will get married?” was used to assess perceived likelihood of marriage. Participants indicated their responses on a 5-point Likert scale. This item is based on an item used in the National Survey of Families and Households. Internal consistency could not be calculated for this single item measure, but test retest reliability has been shown to be high in other research with a similar sample (Rhoades et al., 2010).

Results

Cross-sectional Associations

Table 1 presents the correlations among the four main facets of commitment and other relationship characteristics. Given the large sample size, we set alpha at .01 for these analyses. All facets of commitment were significantly correlated with each other with the exception that felt constraint was not significantly correlated with material constraints nor with perceived constraints (correlations ranged from -.03 to -.63). The highest correlation was between felt constraint and dedication. Higher dedication was associated with lower felt constraint.

Table 1

Correlations among Types of Commitment and Other Relationship Characteristics

With regard to associations with other relationship characteristics, the four facets of commitment were significantly associated with relationship adjustment, perceived likelihood of marriage, perceived likelihood of break up, length of relationship, and actual relationship stability, with three exceptions. The association between material constraints and relationship adjustment was non-significant, as were the associations between dedication and felt constraint with length of relationship. The significant correlations ranged from .18 to -.67 in magnitude.

Felt constraint operated quite differently from the other measures of types of constraint. The associations between felt constraint and relationship adjustment, perceived likelihood of marriage, and perceived likelihood of break-up were all in the opposite direction from the associations among these relationship characteristics and the other three facets of commitment measured. Higher levels of felt constraint were associated with lower relationship adjustment, lower perceived likelihood of marriage, higher perceived likelihood of break-up, and less actual relationship stability, whereas more dedication, more perceived constraints, and more material constraints were associated with higher relationship adjustment, higher perceived likelihood of marriage, lower perceived likelihood of break-up, and less stability. Higher material constraint and perceived constraint commitment were also significantly associated with having been in the relationship longer.

Predicting Break-up

Logistic regression

We used a logistic regression to examine the unique contributions of the four aspects of commitment in predicting break-up (Table 2). Alpha was set at .01 for these analyses. Dedication as well as perceived, material, and felt constraint scores were entered simultaneously as independent variables. The dependent variable was whether the participants were together with the same person (1) or not (0) by the eight-month follow-up (as described earlier). The results indicated that each facet of commitment explained unique variance in stability; each was significantly related to stability, even when controlling for all of the other measures. Higher levels of dedication, perceived constraints, and material constraint, and lower levels of felt constraint, all uniquely predicted staying together.

Table 2

Summary of Logistic Regression Predicting Stability from Four Aspects of Commitment

Sensitivity analyses

To determine whether it would be important to include interactions between gender and commitment variables or cohabitation status and commitment variables in predicting break-up, we first examined interaction effects at the bivariate level. Although there was a main effect of gender for one variable (women reported significantly higher dedication than men), there were no significant interactions between gender and later relationship status on any of the commitment variables in the bivariate analyses, so interactions between commitment variables and gender were not included in the main logistic regression models. With regard to cohabitation status, there were main effects of cohabitation status for perceived, material, and felt constraints, with those cohabiting reporting more constraints than individuals who were not cohabiting, and there was one significant interaction, between relationship status and cohabitation status for material constraints. This interaction indicated that the difference between those who were still together and those who were not was larger for those who were living together compared to those who were dating. That is to say, material constraints were more predictive of stability for those who were living together than for those who were not. Based on these analyses, we included the interaction term for material constraints by cohabitation status in the initial logistic regression we ran to test whether cohabitation status moderated the association between material constraints and relationship outcome. When controlling for the other facets of commitment, however, the interaction between material constraints and cohabitation status was not significant, so we did not include it in the final logistic regression model.

We also conducted a logistic regression with the four facets of commitment predicting break-up while controlling for several demographic variables: length of relationship, parental status (couple has a biological child together, participant has his or her own child from previous relationship, and/or partner has a child from a previous relationship), education, income, race (African American, Asian, or other with White as the reference group), and ethnicity (Latino vs. non-Latino). The results of this logistic regression indicated that even when controlling for these demographic factors, each type of commitment uniquely predicted relationship stability. Detailed results of these sensitivity analyses can be obtained from the first author.2

Discussion

This study examined four different facets of commitment in a large sample of individuals who were dating exclusively, but not yet married. These four facets (dedication and perceived, felt, and material constraints) were generally associated cross-sectionally with other relationship characteristics, including relationship adjustment, perceived likelihood of break-up and perceived likelihood of marriage. Felt constraint was negatively related to relationship adjustment while the other facets of commitment were positively related. Next, we tested hypotheses about how these four facets of commitment were related to relationship stability over an eight-month period (a period in which 26.3% of the relationships ended). In support of our prediction, each of the four facets uniquely predicted relationship termination. Less dedication, less perceived constraint commitment, fewer material constraints, and a stronger sense of felt constraint each predicted break-up, collectively explaining 16% of the variability in break-up.

Before considering the implications of our findings and how they relate to prior research in the field, let us consider this study’s limitations. First, although most of the measures we used had been used in previous research, our measures of material and felt constraints were novel and have not been tested in other published research. These measures demonstrated high internal consistency in this study and were related to other constructs in predictable ways, but more work could be done to establish their validity. Second, the sample does not include any married participants, and the nature and function of constraints in marriage may be different from dating and cohabiting relationships. Indeed, we believe part of the value of this work lies in the fact that the field knows so little about relationship stability and dissolution in romantic relationships prior to marriage. Hence, with these limitations in mind, the current study extends knowledge of how commitment, and particularly different kinds of constraints, is related to relationship stability in unmarried romantic relationships.

As mentioned earlier, another recent study examined how different kinds of constraints (referred to as investments) were associated with relationship termination (Goodfriend & Agnew, 2008). Of the four kinds of investments tested in this prior research, what Goodfriend and Agnew called past tangible investments are most similar to the material constraints measured here. Our results are different from these previous findings in that material constraints were cross-sectionally related to dedication and predictive of relationship continuance whereas past tangible investments were not significantly related to either construct in Goodfriend and Agnew’s study. Differences in the samples (college students vs. national and diverse) and in item scaling (continuous vs. dichotomous) may account for the differences in findings. Our research highlights the importance of continuing to consider material constraints (or past tangible investments) in future work, as they predicted staying together, independent of dedication and other forms of constraint.

Along these lines, it is remarkable that material constraints, such as sharing debt or owning a pet, are associated with staying in a relationship, even controlling for one’s desire to maintain the relationship (dedication), one’s sense of feeling trapped or not (felt constraint), and the constraints that he or she is aware of (perceived constraints). Can sharing a gym membership or adopting a cat really make a person stay in a relationship? Our results indicate that adding just one additional material constraint is associated with a 10% increase in the odds of staying together, even when controlling for the other aspects of commitment. One explanation for this finding comes from Stanley, Rhoades, and Markman (2006) who suggested that constraints creep up on cohabiting couples, creating an inertia that favors relationship continuance regardless of the quality of the match between partners, which in turn reduces options and alternatives before partners realize it. Many of the small material constraints measured here may not be recognized as events or actions that could make it harder to break up. On the one hand, these small material actions likely occur, in part, because of growing dedication between partners (Stanley & Markman, 1992), and in cases in which partners are already highly dedicated, material constraints are unlikely to be troublesome. That is, they will not actually feel “constraining” unless one wishes the relationship would end. On the other hand, these material constraints predicted stability even when controlling for dedication, suggesting that they play at least an incremental independent role in relationship continuance. Future studies might examine which material constraints are the most potent predictors of relationship outcomes. Such work could lead to a greater understanding of which types of behavior in developing relationships are most likely to reduce one’s options for finding a good relationship match before dedication is fully developed and clarified. There are likely some material constraints that are more important than others and the current study included only a summed score.

Our findings also indicated some variation in commitment patterns based on type or stage of relationship. There was preliminary evidence that the association between material constraints and relationship stability might be stronger for couples who are cohabiting compared to those who are not, however this was only true in the bivariate analyses, not when controlling for other facets of commitment. This finding overlaps with some theory suggesting that it is likely more difficult to terminate cohabiting relationships than to end non-cohabiting relationships, partly due to constraints that increase when partners begin living together (Stanley, Rhoades, & Markman, 2006). It could also be related to the fact that many individuals in dating, non-cohabiting relationship did not report any material constraints (41.5%) at the initial assessment; the lack of variability on this measure could constrain the strength of the association between material constraints and relationship stability among this group. Importantly, we do not suggest that this is an artifact of the procedures or sample; rather, dating relationships likely have very few tangible, material constraints compared to cohabiting relationships.

The finding that felt constraint independently predicts relationship termination is more complex, in some ways, than the other findings. In theory, the concept of constraint implies greater difficulty in ending a relationship, but here, higher felt constraint was associated with higher likelihood of break up. We measured felt constraint using items that reflected a feeling of being trapped. Hence, being trapped reflects both a desire to terminate a relationship but also an awareness of barriers to doing so. A person who is happy in his or her relationship will likely not feel trapped, regardless of various types of constraints. Hence, although felt constraint likely slows down a break up because it reflects a sense that termination would be emotionally or tangibly taxing, it nevertheless predicts termination because it also reflects strong feelings of wanting out (as it was negatively correlated with relationship adjustment). In this way, this concept of felt constraint may relate to what others have referred to as dissolution consideration, which is an intermediate step between commitment and relationship termination (VanderDrift, Agnew, & Wilson, 2009). Studies with many follow-ups and short intervals between assessments (e.g., a daily diary study) could address these associations between felt constraint, dissolution consideration, and break-up more directly.

More generally, future research with longer term follow-up could test models of how these four aspects of commitment develop over time. For example, it could be that as material constraints increase, perceived constraints also begin to increase. If dedication is high, the development of material constraints and perceived constraints will not feel uncomfortable or be problematic, but if dedication is low and if relationship satisfaction is low, increasing material and perceived constraints could be associated with increases in felt constraint.

Although each of the four aspects of commitment measured here predicted unique variance in relationship stability over time, they accounted for only a small proportion of the variance in short-term break-up behavior. Future research could address additional aspects of commitment as well as other variables, such as communication, violence, or infidelity, that may also help explain why some unmarried relationships persist while others dissolve.

Conclusion

In summary, this study demonstrated in a large, relatively representative sample that both dedication and constraint are related to relationship stability among individuals in unmarried romantic relationships. In addition to measuring perceived constraint commitment, which has been addressed in prior research, we examined two new aspects of constraint commitment: material constraints, such as paying rent together or having purchased a future vacation, and felt constraint, which refers to the sense that one is trapped in a relationship because of internal or external pressures to stay. Along with dedication (the desire to maintain the relationship in the long run), these three aspects of constraint commitment each independently predicted relationship stability. Felt constraint was associated in the opposite direction from the other aspects of commitment measured, but this was not surprising given that it reflects both a sense of wanting out and of feeling trapped.

Commitment can be conceptualized in many ways, but the most fundamental meaning in a romantic relationship is that there is a future. This sense of a future can be based in both a desire for a future with the partner (dedication) and the growing presence of factors that reinforce staying together regardless of that desire (constraints). Taken altogether, commitment should be associated with relationship continuance, and this study demonstrated this association. Since relationships are not continuously satisfying, factors other than mere satisfaction must explain continuance through ups and downs, making attention to factors related to constraint important in the study of romantic relationships.

Acknowledgments

Preparation of this manuscript was supported in part by a grant from The National Institute of Child Health and Human Development (NICHD) to Scott Stanley, Galena Rhoades, and Howard Markman (5R01HD047564). The contents are solely the responsibility of the authors and do not necessarily represent the official views of NIH or NICHD.

Footnotes

1A notable number of participants (n = 114; 9.6%) skipped the item on Dyadic Adjustment Scale that asks them how happy they are in the relationships. The individuals who failed to complete this item were not significantly different from those who completed it on any of the other variables used in the analyses presented here (all ps > .20), so we assume they are missing at random and therefore use pairwise deletion as a method for handling these missing data (see Allison, 2002).

2The ways in which children may be associated with relationship stability were not the focus on the current study, but because children and concern for children’s welfare have been consider potential constraints in prior work (e.g., Stanley & Markman, 1992), we examined how involved children were related to relationship termination. Of the three possible types of involvement (child(ren) belongs to both partners, the participant, or the participant’s partner), only the presence of a child that belonged only to the partner (i.e., partner’s child from previous relationship) was significantly associated with relationship stability, χ2(1, N = 1183) = 11.53, p < .01. Being in a relationship with a partner who had one or more children from previous relationship(s) was associated with a lower likelihood of staying together; 24.1% of those whose partners did not have children broke up while 35.1% of those with partners who had children broke up.

References

• Adams JM, Jones WH. Handbook of interpersonal commitment and relationship stability. Dordrecht, Netherlands: Kluwer Academic Publishers; 1999. [Google Scholar]
• Allison PD. Missing data. Thousand Oaks, CA: Sage; 2002. [Google Scholar]
• Cook KS, Emerson RM. Power, equity and commitment in exchange networks. American Sociological Review. 1978;43:721–739. [Google Scholar]
• Emerson RM. Power-dependence relations. American Sociological Review. 1962;27:31–41. [Google Scholar]
• Goodfriend W, Agnew CR. Sunken costs and desired plans: Examining different types of investments in close relationships. Personality and Social Psychology Bulletin. 2008;34(12):1639–1652. [PubMed] [Google Scholar]
• Homans G. Social behavior as exchange. American Journal of Sociology. 1958;62:597–606. [Google Scholar]
• Impett EA, Beals KP, Peplau LA. Testing the investment model of relationship commitment and stability in a longitudinal study of married couples. Current Psychology. 2001;20(4):312–326. [Google Scholar]
• Johnson MP. Commitment: A conceptual structure and empirical application. Sociological Quarterly. 1973;14(3):395–406. [Google Scholar]
• Johnson MP, Caughlin JP, Huston TL. The tripartite nature of marital commitment: Personal, moral, and structural reasons to stay married. Journal of Marriage and Family. 1999;61(1):160–177. [Google Scholar]
• Kelley HH, Thibaut JW. Interpersonal relations: A theory of interdependence. New York: Wiley; 1978. [Google Scholar]
• Kline GH, Stanley SM, Markman HJ, Olmos-Gallo PA, St Peters M, Whitton SW, et al. Timing Is everything: Pre-engagement cohabitation and increased risk for poor marital outcomes. Journal of Family Psychology. 2004;18(2):311–318. [PubMed] [Google Scholar]
• Kurdek LA. Assessing multiple determinants of relationship commitment in cohabitating gay, cohabitating lesbian, dating heterosexual, and married heterosexual couples. Family Relations. 1995;44(3):261–266. [Google Scholar]
• Le B, Agnew CR. Commitment and its theorized determinants: A meta-analysis of the investment model. Personal Relationships. 2003;10(1):37–57. [Google Scholar]
• Levinger G. Marital cohesiveness and dissolution: An integrative review. Journal of Marriage and Family. 1965;27(1):19–28. [Google Scholar]
• McClelland GH, Judd CM. Statistical difficulties of detecting interactions and moderator effects. Psychological Bulletin. 1993;114:376–390. [PubMed] [Google Scholar]
• Owen J, Rhoades GK, Stanley SM, Markman HJ. The Revised Commitment Inventory: Psychometrics and use with unmarried couples. Journal of Family Issues. in press. [PMC free article] [PubMed] [Google Scholar]
• Ramirez A., Jr An examination of the tripartite approach to commitment: An actor-partner interdependence model analysis of the effect of relational maintenance behavior. Journal of Social and Personal Relationships. 2008;25(6):943–965. [Google Scholar]
• Rhoades GK, Stanley SM, Markman HJ. A Longitudinal investigation of commitment dynamics in cohabiting relationships. Manuscript submitted for publication 2010 [PMC free article] [PubMed] [Google Scholar]
• Rusbult CE. Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology. 1980;16(2):172–186. [Google Scholar]
• Rusbult CE. A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. Journal of Personality and Social Psychology. 1983;45(1):101–117. [Google Scholar]
• Rusbult CE, Martz JM, Agnew CR. The Investment Model Scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships. 1998;5:357–391. [Google Scholar]
• Sabourin Sp, Valois P, Lussier Y. Development and validation of a brief version of the Dyadic Adjustment Scale with a nonparametric item analysis model. Psychological Assessment. 2005;17(1):15–17. [PubMed] [Google Scholar]
• Simon D, Krawczyk DC, Holyoak KJ. Construction of preferences by constraint satisfaction. Psychological Science. 2004;15:331–336. [PubMed] [Google Scholar]
• Spanier GB. Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and Family. 1976;38(1):15–28. [Google Scholar]
• Stanley SM, Markman HJ. Assessing commitment in personal relationships. Journal of Marriage and Family. 1992;54:595–608. [Google Scholar]
• Stanley SM, Rhoades GK, Markman HJ. Sliding vs. deciding: Inertia and the premarital cohabitation effect. Family Relations. 2006;55:499–509. [Google Scholar]
• Stanley SM, Rhoades GK, Whitton SW. Commitment: Functions, formation, and the securing of romantic attachment. Journal of Family Theory and Review. in press. [PMC free article] [PubMed] [Google Scholar]
• Thibaut JW, Kelley HH. The social psychology of groups. Oxford, England: Wiley; 1959. [Google Scholar]
• VanderDrift LE, Agnew CR, Wilson JE. Nonmarital romantic relationship commitment and leave behavior: The mediating role of dissolution consideration. Personality and Social Psychology Bulletin. 2009;35(9):1220–1232. [PubMed] [Google Scholar]