Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Dec 1.
Published in final edited form as: J Fam Psychol. 2010 Dec;24(6):678–687. doi: 10.1037/a0021475

Physical Aggression in Unmarried Relationships: The Roles of Commitment and Constraints

Galena K Rhoades 1,*, Scott M Stanley 1, Gretchen Kelmer 1, Howard J Markman 1
PMCID: PMC3058822  NIHMSID: NIHMS241187  PMID: 21171766

Abstract

Using commitment theory (Stanley & Markman, 1992), the present study explored longitudinal associations between physical aggression and various aspects of commitment and relationship stability. Participants (N = 1278) were unmarried adults between the ages of 18 and 35 who were in a heterosexual romantic relationship at the time of the initial assessment. Of these, 51.6% reported never experiencing physical aggression in their current relationship, 12.8% reported experiencing physical aggression in the past, but not in the last year, and 35.6% reported experiencing physical aggression in the last year. As hypothesized, those who had experienced aggression in the last year were more likely to have broken up one year later. They also generally reported lower levels of dedication and higher levels of constraint commitment compared to those with no history of physical aggression. Lastly, among those who had experienced aggression in the last year, constraints and other commitment-related variables explained more about who broke-up over time than did relationship adjustment alone, indicating the importance of measuring commitment constructs in future research about which aggressive couples are most likely to end their relationships. Clinical implications of these results are discussed, particularly in regard to preventive relationship education programs.

Keywords: Aggression, Commitment, Cohabitation, Romantic Relationship, Violence

Violence in romantic relationships is of significant public health relevance both because of its high prevalence rates and because of its negative consequences. Prevalence rates of aggression are difficult to estimate, but there is evidence across a range of samples that nearly half of all couples experience some form of physical aggression at some point in their relationship (Lawrence & Bradbury, 2001; Pedersen & Thomas, 1992; Slep & O’Leary, 2005). It is well-documented that physical aggression between partners is associated with a host of negative outcomes for both the adults and children involved, ranging from mental and physical health problems to reduced work productivity and cognitive abilities (e.g., Knickerbocker, Heyman, Slep, Jouriles, & McDonald, 2007; Leone, Johnson, Cohan, & Lloyd, 2004). Taken together, these data indicate the importance of understanding the mechanisms that may contribute to relationships with aggression being sustained over time.

Prior research has frequently focused on understanding why women stay with batterers. Some of the key issues influencing the decision to stay include a lack of community resources and support, poverty, negative beliefs about oneself, hope for recovery, and fear of retaliation (Barnett, 2001; Walker, Logan, Jordan, & Campbell, 2004). The current study was not focused on severe interpersonal violence, but rather on more common physical aggression, or what Johnson and Ferraro (2000) call situational couple violence. We used the framework of commitment theory to elucidate processes that help explain why some unmarried relationships that are aggressive continue while others end. Using a national sample of 18–35 year-old men and women in dating or cohabiting relationships, we examined how physical aggression in relationships was associated concurrently with several indices of commitment and prospectively with relationship stability over a twelve-month period.

The existing literature on aggression and commitment in romantic relationships is sparse. Some research shows that physical aggression is associated with a higher likelihood of relationship termination over time (Carlson, McLanahan, & England, 2004; Lawrence & Bradbury, 2001; Shortt, Capaldi, Kim, & Owen, 2006), suggesting that aggression might be associated with lower relationship commitment, yet other studies have concluded that aggression is associated with higher commitment (Hammock & O’Hearn, 2002; Pedersen & Thomas, 1992). In these previous studies that used the term commitment, however, it was operationalized merely as relationship status, such as being married or not. As others have suggested, distinguishing among different facets of commitment may be particularly important when examining physical aggression in relationships (Frye, McNulty, & Karney, 2008) because simple categorizations of commitment (such as marital status) do not distinguish between a desire to stay and a sense of obligation to stay. Thus, for the purposes of the current study, we relied on commitment theory to address how specific aspects of commitment may be important in understanding the association between aggression and relationship stability.

Based in part on interdependence (Kelley & Thibaut, 1978) and social exchange theories (Cook & Emerson, 1978), theories of commitment generally distinguish between the desire to be in a relationship and forces that compel a couple to stay together or make it more difficult to leave (see Adams & Jones, 1997; Johnson, 1999; Rusbult, 1980). Within Stanley and Markman’s (1992) theory of commitment, the desire to maintain the relationship for the long term is called interpersonal commitment or dedication. Pressures or circumstances that can serve as barriers to relationship termination are referred to as constraint commitment. Constraints can come in many forms, and different types of constraints may impact relationships differently (Goodfriend & Agnew, 2008). Therefore, the current study considered a variety of potential constraints that may serve as barriers to leaving relationships. Specifically, we examined living together, having children together, social pressure from friends and family for the relationship to continue, concern for one’s partner’s welfare in the event of a breakup, perceptions regarding the quality of alternative life choices, structural and material investments that would be lost if the relationship ended, perceptions regarding the potential difficulty of terminating the relationship, the availability of suitable alternative partners, the length of the relationship, and the sense that one is trapped in the relationship. Theoretically, these kinds of constraints explain why some relationships continue even though they are not particularly satisfying or when dedication is low (Stanley & Markman, 1992). Hence, constraints could help explain why people remain in aggressive relationships. Although previous research has established a negative association between physical aggression and general relationship quality (McKenry, Julian, & Gavazzi, 1995; Leonard & Blane, 1992; Katz, Washington Kuffel, & Coblentz, 2004), no research has tested how aggression is related to these specific indices of constraint commitment described above. A better understanding of the association between these types of constraints and aggression could inform both our knowledge of the complex motivations involved in stay-leave decisions and how best to address violence in prevention and intervention programs.

Present Study

The purpose of this paper was to investigate how experiences of physical aggression in one’s current relationship were related to aspects of commitment and relationship stability over time. Specifically, we tested how having experienced physical violence in the current relationship was related concurrently to several indices of commitment and to the likelihood of being together twelve months later. We divided participants into three groups based on their history of aggression in the current relationship: 1) those who reported no physical aggression ever in the current relationship, 2) those who experienced physical aggression in the last year, and 3) those who experienced physical aggression at some point in the past (with their current partner) but not within the last year. We hypothesized that having a history of physical aggression in the current relationship, particularly within the last year, would be associated with a higher likelihood of break-up as well as with lower dedication and more constraints.

There is an apparent contradiction in the expectation that relationships with a history of aggression would be both more likely to break up and characterized by more constraints. Aggression tends to be associated with lower satisfaction (e.g., Katz et al., 2004) and therefore would be expected to predict ending the relationship. At the same time, commitment theory suggests that satisfaction is not the only reason partners stay together. Constraints or investments in the relationship can also serve as barriers to ending the relationship, even when satisfaction or dedication is low (Rusbult, 1980; Stanley & Markman, 1992). We predict that constraints may help explain why relationships with aggression are intact. To examine this possibility prospectively, we tested the hypothesis that among those who had experienced aggression in the last year, commitment-related constructs would explain additional variance in relationship stability over time, over and above relationship adjustment. Support for this hypothesis would highlight the importance of considering commitment, particularly constraint commitment, in understanding stay-leave behavior among those in relationships with aggression.

We did not predict gender differences in the way physical aggression would be related to relationship stability or indices of commitment, however, gender differences have often been a focus in research on violence (see Katz, Kuffel, & Coblentz, 2002; Langhinrichsen-Rohling, in press; Ross & Babcock, in press). Thus, we also tested for gender moderation in this study.

Method

Participants

Participants (N = 1278) in the current study were individuals who took part in the first three waves of a larger, longitudinal project on romantic relationship development (Rhoades, Stanley, & Markman, in press). The current sample included 468 men (36.6%) and 810 women. At the initial wave of data collection, participants ranged in age from 18 to 35 (M = 25.58 SD = 4.80), had a median of 14 years of education and a median annual income of $15,000 to $19,999. All participants were unmarried but in romantic relationships with a member of the opposite sex. At the initial assessment, they had been in their relationships for an average of 34.28 months (Mdn = 24 months, SD = 33.16); 31.9% were cohabiting. In terms of ethnicity, this sample was 8.2% Hispanic or Latino and 91.8% not Hispanic or Latino. In terms of race, the sample was 75.8% White, 14.5% Black or African American, 3.2% Asian, 1.1% American Indian/Alaska Native, and 0.3% Native Hawaiian or Other Pacific Islander; 3.8% reported being of more than one race and 1.3% did not report a race. With regard to children, 34.2% of the sample reported that there was at least one child involved in their romantic relationship. Specifically, 13.5% of the sample had at least one biological child together with their current partner, 17.1% had at least one biological child from previous partner(s), and 19.6% reported that their partner had at least one biological child from previous partner(s). The larger study included 1293 participants, but there were 15 individuals who were missing data on physical aggression. These individuals were therefore excluded from the current study, leaving a final N of 1278.

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, 154 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 1293 for the first wave (T1) of data collection. These 1293 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 and the fourth wave (T4) was mailed four months after T3. Data from T2, T3, and T4 were only used for measuring relationship stability (described below).

Measures

Demographics

Several items were used to collect demographic data, including age, ethnicity, race, income, and education. Others were used to determine the length of the current relationship, whether the couple was living together (“Are you and your partner living together? That is, do you share a single address without either of you having a separate place?”), and whether the couple had children together and/or by previous partners.

Physical aggression

Certain subscales of the Revised Conflict Tactics Scale (Straus, Hamby, Boney-McCoy, & Sugarman, 1996) were given to all participants. We used the minor injuries received subscale (e.g., “I had a sprain, bruise, or small cut because of a fight with my partner”) as well as the minor physical aggression toward partner (e.g., “I pushed or shoved my partner”) and by partner subscales (e.g., “My partner did this to me”) to create the categorical history of physical aggression variable that was used in the analyses presented in this paper. Individuals who reported that they had never sustained injuries due to a fight with their partner, had never used physical aggression against their partner, and had never been the recipient of physical aggression from their partner were coded as having no history of physical aggression (coded 0). As has been suggested by others (Hanley & O’Neill, 1997), individuals who reported any of these behaviors were coded as either having a history of physical aggression in the last year (coded 2) or having a past history of physical aggression, but not in the last year (coded 1).

Relationship stability

For the current study, relationship stability data were obtained from T2, T3, and T4. Individuals who were broken up by either T2, T3, or T4 were included in the broken up group. To be included in the intact group, however, individuals needed to have completed T4 and indicated then that they were in the same relationship as at T1. We made these inclusion and coding rules so that we could compare those who broke-up within the year following T1 to those who stayed together during for this entire year period. Of the initial sample, 1060 (82.9%) participants met criteria for one of these groups. Of the final 1060 who were used in the relationship stability analyses, 736 (69.4%) remained in the same romantic relationship over the twelve-month period and 324 (30.6%) had broken up within the twelvemonth timeframe. Chi-square analyses indicated no significant (p > .20) associations between missingness on relationship stability and any other study variables, with the exception of the alternative quality scale. Those who were missing relationship stability data reported lower quality alternatives to the relationships at T1 than those who were not missing data.

Dedication

Dedication was measured with the 14 items from the Commitment Inventory that are designed to measure dedication (Stanley & Markman, 1992). The Commitment Inventory was originally developed using novel items and constructs as well as constructs from Johnson’s work on commitment (Johnson, 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). 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” and “I like to think of my partner and me more in terms of ‘us’ and ‘we’ than ‘me’ and ‘him/her.’” 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.

Constraints

To measure potential constraints, we used several items and scales. First, we assessed whether participants were living with their partners using the item, “Are you and your partner living together? That is, do you share a single address without either of you having a separate place? (no = 0, yes = 1). Second, we asked whether participants had biological children with their current partner (no = 0, yes = 1) and/or by previous partners (no = 0, yes = 1).

Third, we used six subscales from the Commitment Inventory (Stanley & Markman, 1992) to assess perceived constraints. These subscales measure social pressure (4 items, α = .77, e.g., “It would be difficult for my friends to accept it if I ended the relationship with my partner”), concern for partner’s welfare (3 items, α = .48, e.g., “I could not bear the pain it would cause my partner to leave him/her even if I really wanted to”), alternative quality of life (5 items, α = .66, e.g., “I would not have trouble supporting myself should this relationship end (reverse-coded)”), structural investments (4 items, α = .68, e.g., “I have put a number of tangible, valuable resources into this relationship”), termination procedures (3 items, α = .79, e.g., “The steps I would need to take to end this relationship would require a great deal of time and effort”), and availability of alternative partners (4 items, α = .63, e.g., “I believe there are many people who would be happy with me as their spouse or partner (reverse-coded)”). The reliability and validity of these subscales have recently been demonstrated in unmarried samples (Owen, Rhoades, Stanley, & Markman, in press). In the same study, a confirmatory factor analysis supported the validity of measuring each area of constraint commitment separately.

Fourth, to measure material constraints, we used The Joint Activities Checklist (Rhoades, Stanley, & Markman, 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.

Lastly, we measured felt constraint using three items that measure how constrained one feels in a relationship: “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. This scale has been shown to be reliable and related to other constructs in expected directions in other research (author citation; Stanley, Markman, & Whitton, 2002). The mean of the three items was used for analyses with higher scores indicating feeling more trapped or stuck, α = .82.

Relationship adjustment

To assess global relationship adjustment, we used the 4-item version of the Dyadic Adjustment Scale (Sabourin, Valois, & Lussier, 2005; Spanier, 1976). This brief version of the original 32-item measure has been shown to be internally consistent, highly correlated with the original, and a valid predictor of relationship stability over time (Sabourin et al, 2005). The four items tap relationship happiness, whether a couple has considered separation, a general sense that the relationship is going well, and how often partners confide in one another. The total score was used, with higher scores reflecting higher relationship adjustment, α = .80.

Results

Descriptive Findings

In this sample, 51.6% reported never experiencing physical aggression in their current relationship, 12.8% reported experiencing physical aggression at some point in their current relationship, but not in the past year, and 35.6% reported experiencing physical aggression in the last year. Of those who had experienced aggression in the last year, 36.0% reported that they had experienced physical pain the next day or a sprain, bruise, or small cut at least once after an aggressive episode with their partner. Of those who had experienced aggression in the past, but not in the last year, 15.3% reported having had sustained these kinds of injuries. In terms of frequency of aggressive acts, among those who had experienced aggression in the past year, 28.1% reported that either they or their partner had grabbed, pushed/shoved, slapped, thrown something at, or twisted the arm or hair of the other partner 6–10 times or more during the past year. Thus, the majority of participants who had experienced aggression in the last year had not sustained injuries and experienced aggression less than monthly, on average. Data on the frequencies of these behaviors for those who had experienced aggression at some point in their current relationship, but not in the past year, were not available.

We next tested whether there were differences between men and women in reports of aggression in the past, in the last 12 months, or not at all. There were no significant gender differences in terms of prevalence.

Correlations among the dependent variables in Table 1 ranged in absolute strength from .01 (for living together and unavailability of other partners) to −.64 (for felt constraint and dedication). The median absolute value of the correlations among dependent variables was .16; the average was .17 and only three correlations were .4 or above, indicating that most variables were not very highly correlated and that they tend to measure different constructs.

Table 1.

Differences in Commitment Indices based on Relationship Aggression History

Aggression History No/Past Past/Last No/Last Omnibus Test
None n = 660 Past n = 163 Last Year n = 455
ANOVAs
Construct M (SD) M (SD) M (SD) Effect Sizes
Relationship length 29.05 (29.84) 52.44 (41.11) 35.38 (32.26) −0.66*** 0.47*** −0.20** F(2, 1274) = 34.47***
Dedication 5.46 (0.99) 5.41 (0.93) 5.15 (1.08) 0.05 0.26** 0.30*** F(2, 1273) = 13.08***
Felt constraint 1.83 (1.16) 2.15 (1.41) 2.58 (1.56) −0.25* −0.29*** −0.56*** F(2, 1272) = 42.22***
Alternative quality 2.36 (1.13) 2.64 (1.19) 2.57 (1.22) −0.24* 0.05 −0.18** F(2, 1273) = 6.20**
Unavail. of others 2.79 (1.17) 2.74 (1.22) 2.77 (1.16) 0.04 −0.02 0.02 F(2, 1273) = 0.12
Concern for partner 4.04 (1.34) 4.17 (1.23) 4.17 (1.37) −0.10 0.00 −0.10 F(2, 1273) = 1.67
Social pressure 4.06 (0.86) 4.12 (0.85) 3.91 (0.89) −0.08 0.24* 0.17* F(2, 1273) = 5.26**
Structural invest. 3.63 (1.28) 4.16 (1.29) 4.14 (1.32) −0.41*** 0.02 −0.39*** F(2, 1271) = 24.58***
Term. procedures 4.40 (1.70) 5.06 (1.51) 4.84 (1.61) −0.41*** 0.14 −0.27*** F(2, 1270) = 15.76***
Material constraints 2.74 (3.65) 4.64 (4.26) 3.21 (3.40) −0.48*** 0.38*** −0.13 F(2, 1275) = 17.90***
Chi-Square Analyses
Percent who… % (n) % (n) % (n) Subgroup Differences
Live together 25.5 (168) 50.3 (82) 34.7 (158) *** *** *** χ2(2, N = 1278) = 39.70***
Have child together 7.3 (48) 24.5 (40) 18.5 (84) *** ns *** χ2(2, N = 1278) = 48.65***
Have prev. child 16.7 (110) 17.8 (29) 17.6 (80) ns ns ns χ2(2, N = 1278) = 0.22
Open in a new tab

Notes.

***

p < .0005,

**

p < .01,

*

p < .05.

Effect sizes are Cohen’s d values. ANOVAs were used to test differences among the history of aggression groups on continuous measures and means and standard deviations are presented for those analyses. Chi-square tests were used for categorical variables and percentages and subgroup ns are for those analyses.

Physical Aggression and Relationship Stability (Hypothesis 1)

We predicted that having experienced physical aggression would be associated with a higher likelihood of relationships ending. We tested this prediction using a logistic regression with dummy variables for physical aggression in the past, but not in the last year (0 = no, 1 = yes) and aggression in the last year (0 = no, 1 = yes) were entered as predictors of later relationship status (0 = broke-up, 1 = together). The results indicated that, compared to those with no history of aggression, those who experienced aggression in the last year were significantly less likely to be together twelve months later (B = −.46, SE = .15, Wald = 9.97, Odds ratio = .63, p < .01). In contrast, violence in the past but not the last year was not a significant predictor of later relationship status. Thus, our hypothesis regarding stability was supported; 26.7% (n = 149) of those with no history of aggression, 30.4% (n = 41) of those who had experienced aggression in the past, but not in the last year, and 36.5% (n = 134) of those who experienced aggression in the last year broke up. A follow-up logistic regression with gender X aggression history interactions indicated that this finding was not moderated by gender.

Physical Aggression and Indices of Commitment (Hypothesis 2)

We predicted that having a history of physical aggression would be related to lower dedication and more barriers to leaving (or constraints). To test this prediction, we examined the bivariate relationships between indices of commitment and having never experienced aggression in the current relationship, having experienced aggression at some point, but not in the past year, or having experienced aggression within the last year. Results from these one-way ANOVAs and chi-squares are presented in Table 1. The findings indicated several significant differences between those who had never experienced aggression and those who had experienced aggression within the last year. Those with no history of aggression were more dedicated, experienced less felt constraint, and reported fewer constraints in terms of alternative quality of life, structural investments, and termination procedures, as well as more social pressure to stay together. They also had been in their relationships for a shorter period of time, were less likely to have children with their partners, and were less likely to be cohabiting compared to those who had experienced aggression in the past year. Similar differences existed between those with no aggression and those who had experienced aggression, but not in the last year. Specifically, those with no aggression reported less felt constraint, fewer material constraints, shorter relationship duration, a lower likelihood of having children together and of living together as well as fewer constraints in terms of alternative quality of life, structural investments, and termination procedures than those with a past history of aggression. Lastly, significant differences also existed between those who had experienced aggression within the last year and those who had experienced it, but not in the past year; those with recent experience had lower dedication, higher felt constraint, less social pressure to stay together, as well as fewer material constraints and shorter relationship duration. Thus, overall, our hypotheses about differences on commitment indices based on history of physical aggression were supported. A history of aggression was generally associated with less dedication and more constraints. The largest differences appeared to between those with no history of aggression and those who had experienced aggression in the last year.

To check for gender moderation for the indices of commitment and their association with history of aggression (hypothesis 2), we conducted 2(gender) X 3(aggression history) ANOVAs. There was only one variable for which the interaction between gender and aggression history was significant: having children together, F(2,1272) = 5.23, p < .01. Because this variable is nominal in nature, we probed this interaction by running chi-square tests separately for men and women. For men, having children together was not significantly related to reports of aggression, but for women, those who had never experienced aggression were less likely to have children with their partners (8.2%) than those who reported aggression in the past, but not the last year (30.9%), and those who reported aggression within the last year (23.1%).

Predicting Break-up Among those who Experienced Aggression in the Last Year (Hypothesis 3)

Our final hypothesis was that among those who had experienced aggression in the last year, commitment-related constructs would explain additional variance in relationship stability over time, controlling for relationship adjustment. To test this hypothesis, we compared the variance explained in relationship stability from two logistic regressions. The first logistic regression included only relationship adjustment as a predictor of relationship stability over the 12-month period. According to the Nagelkerke R2 statistic, relationship adjustment alone explained 9.1% of the variance in relationship stability, B = 0.14, SE(B) = 0.03, eB = 1.16. Our second logistic regression (Table 2) included relationship adjustment plus all of the commitment-related variables measured here (i.e., dedication, felt constraint, alternative quality, unavailability of other partners, concern for partner’s welfare, social pressure to stay together, structural investments, termination procedures, material constraints, living together status, length of relationship, and having a child together or by a previous partner). This model explained 31.9% of the variability in relationship stability. A chi-square test comparing the log-likelihoods for these two models confirmed our hypothesis that the model including the indices of commitment would explain additional variance in relationship stability over the relationship-adjustment only model, χ2(13) = 128.51, p < .001. In separate logistic regressions, relationship adjustment, living together, longer length of relationship, higher dedication, lower felt constraint, lower alternative quality, more unavailability of other partners, higher social pressure to stay together, more structural investments, greater difficulty in termination procedures, more material constraints, and having a child together each predicted staying together (significant Wald-test values ranged from 4.23 to 28.45, ps < .05) but when entered simultaneously in the model in Table 2, only relationship adjustment, living together, length of relationship, social pressure, and the unavailability of other partners were unique predictors of stability.

Table 2.

Predicting Relationship Stability Among those with Aggression in the Last Year

Predictor B SE (B) eB
Relationship adjustment 0.15** 0.05 1.16
Living together 1.59*** 0.41 4.92
Relationship length 0.01* 0.01 1.01
Child together 0.43 0.47 1.54
Child from previous rel. −0.32 0.39 0.73
Dedication 0.04 0.19 1.05
Felt constraint 0.15 0.14 0.86
Alternative quality −0.04 0.19 0.96
Unavail. of others 0.29* 0.14 1.33
Concern for partner −0.21 0.11 0.81
Social pressure 0.36* 0.17 1.43
Structural invest. 0.03 0.12 1.03
Term. procedures 0.15 0.09 1.16
Material constraints −0.08 0.06 0.93
Constant −4.15
Model fit statistics χ2(14, N = 326)= 64.30***
Open in a new tab

Notes.

***

p < .001,

**

p < .01,

*

p < .05.

SE (B) = standard error, eB = exponentiated B.

Individuals who stayed together were coded as 1, individuals who broke up were coded as 0.

To check for differences between men and women in the variance explained in break-up behavior among those who had experienced aggression in the last year (hypothesis 3), we examined the two logistic regressions described above among men and among women separately. The results were very similar across men and women, with the full logistic regressions (including all commitment-related variables) explaining two to three times more variance in relationship stability than the model that included only relationship adjustment.

Discussion

In this large, national sample of unmarried young adults in opposite-sex relationships, nearly half of them (48.4%) reported experiencing some sort of physical aggression in the current relationship. The findings show that having a history of aggression in this relationship was related in predicted directions to several different aspects of relationship commitment and also with relationship stability. Further, findings from the current study demonstrate that among those who have a history of aggression in the last year, constraints explain additional variability in who breaks up than relationship adjustment alone. Thus, considering commitment-related constructs seems to be important in knowing which couples with a history of aggression will stay together or break-up.

Previous research on physical aggression and commitment has indicated a positive association in which couples who experience aggression tend report higher levels of commitment (Hammock & O’Hearn, 2002; Pedersen & Thomas, 1992), but these studies have typically conceptualized commitment as marital status. In contrast, we used commitment theory (Stanley & Markman, 1992) as a framework to study the associations between aggression and dedication and constraint and found that having experienced aggression in the last year was associated with less dedication to the relationship, but also more barriers to leaving, compared to those who did not have a history of aggression. Specifically, those who had experienced aggression in the last year reported more material constraints, such as sharing financial obligations, as well as more perceived constraints, such as believing other suitable partners are not available or experiencing social pressure to stay together. In addition, they tended to have been in their relationships longer and were more likely to be living together and to have a child together than those without a history of aggression. When dedication is low and constraint commitment is high, it should be associated with discomfort and unease for the involved partners (Stanley & Markman, 1992); in the current study, this discomfort is evidenced by the finding that those who had experienced aggression in the last year were more likely to report feeling trapped than those who had not.

There were fewer differences between those without a history of physical aggression and those who experienced it only in the past but not in the last year. Those who had experienced aggression in the past, but not in the last year, had more barriers to leaving and reported feeling more trapped than those with no history of aggression. These differences tended to be smaller in size than the differences between those who had experienced aggression in the last year and those who had not experienced any aggression in the current relationship. These findings suggest that those who had experienced aggression previously, but not in the past year, may have made gains in terms of relationship safety compared to an earlier period in the relationship.

Although not hypothesized, there were also some differences in commitment between those who had experienced aggression in the last year and those who had experienced it only in the past. Those with aggression in the last year had lower dedication, and interestingly, fewer material constraints than those with a past history of physical aggression. Given the general patterns of findings across the three types of aggression history, and particularly the differences between those who had experienced aggression in the past, but not in the last year, these distinctions seem meaningful. These groups will be fluid over time, as some who have not experienced aggression may experience it in the future, some who experienced it only in the past will again experience aggression in the future, and some who experienced it recently will later move into the “only in the past” group. It is also the case that those who experienced aggression only in the past reported less injury, so these findings may reflect something about the severity of aggression in addition to the timing of the aggression. Future research could more qualitatively describe the distinctions based on history of aggression. For example, it might be particularly important in terms of clinical practice to know more about couples who experienced aggression at some point but were able to stop this negative pattern of interaction and continue the relationship. How did these unmarried couples change their negative behaviors? Are these lasting changes?

The current study can speak to how having experienced physical aggression is associated with relationship stability over time. Although a more distal history of aggression was not significantly related to breaking up, individuals who experienced aggression in the last year were more likely to end their relationships than those with no history of aggression at all. These findings are in line with other research showing an association between aggression and risk for relationship dissolution (Heyman, O’Leary, & Jouriles, 1995; Rogge & Bradbury, 1999; Shortt et al., 2006), though we examined a shorter time frame (twelve months) than previous research had. The current findings extend the previous literature by showing that the relationships that had experienced aggression in the last year were also characterized by more constraints and barriers to leaving. This combination of findings suggests that ending the relationship may have been more difficult, perhaps both emotionally and logistically, for those who had experienced aggression compared to those who had not. A wide literature exists on the process of separation from relationships, particularly marriages, characterized by battering and control (see Walker et al., 2004 for a review), but less is known about unmarried separations, especially for those with lower levels of physical aggression. The results from the current study indicate that dedication and constraints help explain more than 30% of the variability in break-ups among those who had experienced aggression in the last year, suggesting the utility of continuing to measure commitment in research on physical aggression and relationship stability.

Several of the commitment constructs individually predicted staying together, but when they were all entered simultaneously, few remained unique predictors of stability, likely because they overlap to various degrees in terms of their meaning. Future research could explore the relationship dissolution process in aggressive relationships more qualitatively to better understand how specific constraints or barriers might influence separation decisions as well as the potential distress associated with a break-up.

One particularly important constraint to consider in future work may be cohabitation. Our findings indicated that those who were living together were more likely to have experienced physical aggression (58.8%) than those who were dating and not living together (43.4%). Among those who had experienced aggression in the last year, living together was also a strong predictor of remaining in the relationship over time. Although few studies have compared the rates of aggression between cohabiting and dating relationships, those that have also indicate a higher prevalence of physical aggression in cohabiting relationships (Brown & Bulanda, 2008; Kline et al., 2004; Magdol, Moffitt, & Caspi, 1998). Additionally, the literature indicates that cohabiting couples experience more aggression than married couples (Brown & Bulanda, 2008; Stanley, Whitton & Markman, 2004) and that this difference is at least partially accounted for by social isolation (Stets, 1991). Similar processes may explain the higher prevalence of aggression in cohabiting relationships as compared to dating relationships in the current study. It may be that cohabitation represents a relationship type or stage in which partners have relatively less social support and more social isolation, making violence more likely to occur and also less likely to be recognized by friends, family members, or professionals who would otherwise intervene in the relationship. Cohabiting couples also tend to have been dating longer and thus they have had more time to experience aggression. Additionally, they likely have more day-to-day contact with each other than dating couples and therefore more opportunities for physical aggression to occur (Magdol et al., 1998). Cohabiting relationships tend to be more constraining than dating relationships (Rhoades et al., 2010), which may also make them more difficult to end even if aggression occurs. Unfortunately, there is evidence that the higher rates of aggression among cohabiting couples continues into marriage, as premarital cohabitation is a risk factor for experiencing physical aggression in marriage, as well (Stanley et al., 2004).

Limitations and Future Directions

This study had several strengths, particularly with regard to the generalizability of the sample and in terms of the in-depth measurement of commitment, but there were also limitations that should be considered. First, because of the nature of our research questions, we defined a history of physical aggression in a dichotomous way; future work could expand the measurement of physical aggression to examine severity or types of violence more comprehensively. Others have detailed that different forms and severities of family violence should be considered separately (e.g., Emery & Laumann-Billings, 1998; Johnson & Ferraro, 2000; Kitzmann, Gaylord, Holt, & Kenny, 2003). In our sample, the majority of individuals appear to have experienced relatively infrequent aggression that did not cause injury, thus our findings may not generalize to couples who experience the most severe kinds of relationship violence. Second, some of the perceived constraint scales had low internal consistency, which may have limited statistical power to detect to differences among the groups on these scales. Third, some research has indicated that the ways in which constraints affect verbal aggression in relationships may depend in part on the other partner’s sense of constraints (Frye et al., 2008). Thus, future research on commitment and physical aggression could benefit from measuring both partners in a dyad. Lastly, given that physical aggression and commitment were measured at the same time point, we are unable to discern the directionality of the findings regarding these variables. We have assumed that the constraints measured here co-vary with aggression, but that they are not necessarily causally related to aggression. On the other hand, resource theory would suggest that couples who are more constrained might evidence higher levels of physical aggression because they have fewer resources or means for handling conflict well or getting help (Atkinson, Greenstein, & Lang, 2005; Goode, 1971). More in-depth, longer-term longitudinal research could better disentangle the relationships among dedication, constraints, aggression, and relationship stability. Such work could also help us understand patterns of break-ups and reunions over time and how these patterns may be related to physical aggression.

Clinical Implications

The results of the current study suggest that violent relationships are characterized by lower levels of dedication and higher levels of constraint commitment. In interventions with individuals who are in relationships that include physical aggression, explaining types of commitment could be a useful way to help individuals make clearer decisions about whether to stay in their relationships or leave them. For example, understanding the difference between dedication and constraint commitment could help an individual recognize whether they want to stay or feel they must stay. Further, identifying the specific barriers and constraints in relationships could be a way to help individuals in unhealthy relationships consider options for mitigating the costs of leaving and ultimately exiting the relationship safely.

Although several specific violence prevention programs exist for adolescents or college students, few are available to those who are older or not in school (Cornelius & Resseguie, 2007). More broadly, others have noted that relationship education efforts geared toward helping individuals or couples improve their relationships and maintain them over time rarely incorporate information on aggression (Lawrence & Bradbury, 2001). One exception is a new program that includes a strong focus on recognizing and preventing violence in relationships, Within My Reach (Pearson, Stanley, & Kline, 2005). As tested with a sample of at-risk women with low-income levels, preliminary results regarding the effectiveness of this program indicate positive increases in communication and relationship quality over time, as well as a trend toward reducing physical aggression (Antle et al., in press). The results of the current study, especially regarding the high prevalence of aggression in dating and cohabiting relationships, bolster the importance of relationship education programs like this one that address violence directly.

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

Publisher's Disclaimer: The following manuscript is the final accepted manuscript. It has not been subjected to the final copyediting, fact-checking, and proofreading required for formal publication. It is not the definitive, publisher-authenticated version. The American Psychological Association and its Council of Editors disclaim any responsibility or liabilities for errors or omissions of this manuscript version, any version derived from this manuscript by NIH, or other third parties. The published version is available at www.apa.org/pubs/journals/fam

References

  1. Adams JM, Jones WH. The conceptualization of marital commitment: An integrative analysis. Journal of Personality and Social Psychology. 1997;72(5):1177–1196. [Google Scholar]
  2. Antle BF, Sar BK, Christensen DN, Ellers FS, Karam EA, Barbee AP, et al. The impact of the Within My Reach relationship training on relationship skills and outcomes for low-income individuals. Journal of Marital and Family Therapy. doi: 10.1111/j.1752-0606.2012.00314.x. (in press) [DOI] [PubMed] [Google Scholar]
  3. Atkinson MP, Greenstein TN, Lang MM. For women, breadwinning can be dangerous: Gendered resource theory and wife abuse. Journal of Marriage and Family. 2005;67(5):1137–1148. [Google Scholar]
  4. Barnett OW. Why battered women do not leave, part 2: External inhibiting factors--social support and internal inhibiting factors. Trauma, Violence, and Abuse. 2001;2(1):3–35. [Google Scholar]
  5. Brown SL, Bulanda JR. Relationship violence in young adulthood: A comparison of daters, cohabitors, and marrieds. Social Science Research. 2008;37(1):73–87. [Google Scholar]
  6. Carlson M, McLanahan S, England P. Union formation in fragile families. Demography. 2004;41:237–261. doi: 10.1353/dem.2004.0012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cook KS, Emerson RM. Power, equity and commitment in exchange networks. American Sociological Review. 1978;43:721–739. [Google Scholar]
  8. Cornelius TL, Resseguie N. Primary and secondary prevention programs for dating violence: A review of the literature. Aggression and Violent Behavior. 2007;12(3):364–375. [Google Scholar]
  9. Emery RE, Laumann-Billings L. An overview of the nature, causes, and consequences of abusive family relationships. American Psychologist. 1998;53(2):121–136. doi: 10.1037//0003-066x.53.2.121. [DOI] [PubMed] [Google Scholar]
  10. Frye NE, McNulty JK, Karney BR. How do constraints on leaving a marriage affect behavior within the marriage? Journal of Family Psychology. 2008;22(1):153–161. doi: 10.1037/0893-3200.22.1.153. [DOI] [PubMed] [Google Scholar]
  11. Goode WJ. Force and violence in the family. Journal of Marriage and Family. 1971;33(4):624–636. [Google Scholar]
  12. 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. doi: 10.1177/0146167208323743. [DOI] [PubMed] [Google Scholar]
  13. Hammock G, O’Hearn R. Psychological aggression in dating relationships: Predictive models for males and females. Violence and Victims. 2002;17(5):525–540. doi: 10.1891/vivi.17.5.525.33715. [DOI] [PubMed] [Google Scholar]
  14. Hanley MJ, O’Neill P. Violence and commitment: A study of dating couples. Journal of Interpersonal Violence. 1997;12(5):685–703. [Google Scholar]
  15. Heyman RE, O’Leary KD, Jouriles EN. Alcohol and aggressive personality styles: Potentiators of serious physical aggression against wives? Journal of Family Psychology. 1995;9(1):44–57. [Google Scholar]
  16. Johnson MP. Commitment: A conceptual structure and empirical application. Sociological Quarterly. 1973;14(3):395–406. [Google Scholar]
  17. Johnson MP. Personal, moral, and structural commitment to relationships: Experiences of choice and constraint. In: Adams JM, Jones WH, editors. Handbook of interpersonal commitment and relationship stability. Dordrecht, Netherlands: Kluwer Academic Publishers; 1999. pp. 73–87. [Google Scholar]
  18. Johnson MP, Ferraro KJ. Research on domestic violence in the 1990s: Making distinctions. Journal of Marriage and Family. 2000;62(4):948–963. [Google Scholar]
  19. Katz J, Kuffel SW, Coblentz A. Are there gender differences in sustaining dating violence? An examination of frequency, severity, and relationship satisfaction. Journal of Family Violence. 2002;17(3):247–271. [Google Scholar]
  20. Kelley HH, Thibaut JW. Interpersonal relations: A theory of interdependence. New York: Wiley; 1978. [Google Scholar]
  21. Kitzmann KM, Gaylord NK, Holt AR, Kenny ED. Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology. 2003;71(2):339–353. doi: 10.1037/0022-006x.71.2.339. [DOI] [PubMed] [Google Scholar]
  22. 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. doi: 10.1037/0893-3200.18.2.311. [DOI] [PubMed] [Google Scholar]
  23. Knickerbocker L, Heyman RE, Slep AMS, Jouriles EN, McDonald R. Co-occurrence of child and partner maltreatment definitions, prevalence, theory, and implications for assessment. European Psychologist. 2007;12(1):36–44. [Google Scholar]
  24. Langhinrichsen-Rohling J. Controversies involving gender and intimate partner violence in the United States. Sex Roles (in press) [Google Scholar]
  25. Lawrence E, Bradbury TN. Physical aggression and marital dysfunction: A longitudinal analysis. Journal of Family Psychology. 2001;15(1):135–154. doi: 10.1037//0893-3200.15.1.135. [DOI] [PubMed] [Google Scholar]
  26. Leone JM, Johnson MP, Cohan CL, Lloyd SE. Consequences of male partner violence for low-income minority women. Journal of Marriage and Family. 2004;66(2):472–490. [Google Scholar]
  27. Magdol L, Moffitt TE, Caspi A. Hitting without a license: Testing explanations for differences in partner abuse between young adult daters and cohabitors. Journal of Marriage and Family. 1998;60:41–55. [Google Scholar]
  28. Owen J, Rhoades GK, Stanley SM, Markman HJ. The Revised Commitment Inventory: Psychometrics and use with unmarried couples. Journal of Family Issues. doi: 10.1177/0192513X10385788. (in press) [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pearson M, Stanley SM, Kline GH. Within My Reach. Denver, CO: PREP, Inc; 2005. [Google Scholar]
  30. Pedersen P, Thomas CD. Prevalence and correlates of dating violence in a Canadian university sample. Canadian Journal of Behavioural Science. 1992;24(4):490–501. [Google Scholar]
  31. Rhoades GK, Stanley SM, Markman HJ. A Longitudinal investigation of commitment dynamics in cohabiting relationships. 2010 doi: 10.1177/0192513X11420940. Manuscript submitted for publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rhoades GK, Stanley SM, Markman HJ. Should I stay or should I go? Predicting dating relationship stability from four aspects of commitment. Journal of Family Psychology. doi: 10.1037/a0021008. (in press) [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rogge RD, Bradbury TN. Till violence does us part: The differing roles of communication and aggression in predicting adverse marital outcomes. Journal of Consulting and Clinical Psychology. 1999;67(3):340–351. doi: 10.1037//0022-006x.67.3.340. [DOI] [PubMed] [Google Scholar]
  34. Ross J, Babcock J. Gender and intimate partner violence in the United States: Confronting the controversies. Sex Roles (in press) [Google Scholar]
  35. 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]
  36. 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. doi: 10.1037/1040-3590.17.1.15. [DOI] [PubMed] [Google Scholar]
  37. Shortt JW, Capaldi DM, Kim HK, Owen LD. Relationship separation for young, at-risk couples: Prediction from dyadic aggression. Journal of Family Psychology. 2006;20(4):624–631. doi: 10.1037/0893-3200.20.4.624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Slep AMS, O’Leary SG. Parent and partner violence in families with young children: Rates, patterns, and connections. Journal of Consulting and Clinical Psychology. 2005;73(3):435–444. doi: 10.1037/0022-006X.73.3.435. [DOI] [PubMed] [Google Scholar]
  39. 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]
  40. Stanley SM, Markman HJ. Assessing commitment in personal relationships. Journal of Marriage and Family. 1992;54:595–608. [Google Scholar]
  41. Stanley SM, Markman HJ, Whitton SW. Communication, conflict and commitment: Insights on the foundations of relationship success from a national survey. Family Process. 2002;41(4):659–675. doi: 10.1111/j.1545-5300.2002.00659.x. [DOI] [PubMed] [Google Scholar]
  42. Stanley SM, Whitton SW, Markman HJ. Maybe I do: Interpersonal commitment and premarital or nonmarital cohabitation. Journal of Family Issues. 2004;25(4):496–519. [Google Scholar]
  43. Stets JE. Cohabiting and marital aggression: The role of social isolation. Journal of Marriage and the Family. 1991;53(3):669–680. [Google Scholar]
  44. Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues. 1996;17(3):283–316. [Google Scholar]
  45. Walker R, Logan TK, Jordan CE, Campbell JC. An integrative review of separation in the context of victimization: Consequences and implications for women. Trauma, Violence, and Abuse. 2004;5(2):143–193. doi: 10.1177/1524838003262333. [DOI] [PubMed] [Google Scholar]

RESOURCES