San Diego State University

Library & Information Access

Access to this article has been purchased by San Diego State University Library & Information Access
For more information on other library services, please visit our website: http://library.sdsu.edu

Thank you for using the San Diego State University Library.

Journal of Family Psychology 2009, Vol. 23, No. 2, 236 246

2009 American Psychological Association 0893-3200/09/$12.00 DOI: 10.1037/a0015364

Qualitative Reports of Problems in Cohabiting Relationships: Comparisons to Married and Dating Relationships
Annie C. Hsueh, Kristen Rahbar Morrison, and Brian D. Doss
Texas A&M University
Although previous research has demonstrated increased relationship distress and separation for cohabiting couples, little is known about specic problems cohabiting individuals encounter in comparison to dating and married individuals. This study examines open-ended reports of 1,252 individuals (220 dating, 231 cohabiting, and 801 married) relationship concerns using a detailed, reliable coding system. The top 5 areas considered most problematic by cohabiting individuals were problems in specic areas of their current relationship, individual problems, general communication, arguments, and emotional affection distance. Dating and cohabiting individuals reported similar frequencies of global problems except that cohabiting individuals reported more problems with arguments and fewer problems with relationship commitment. Married and cohabiting individuals had more differences in their reports of relationship concerns; results suggested that cohabiting relationships tended to be both more vibrant and more volatile than marital relationships. However, most differences between relationship types were no longer signicant after controlling for individuals relationship and demographic characteristics. Keywords: cohabitation, dating, marriage, problems, relationships

Cohabitation is becoming increasingly common; in the United States, an estimated 50% (Bumpass & Lu, 2000) to more than 60% (Stanley, Whitton, & Markman, 2004) of couples live together before getting married. Despite the increasing prevalence of cohabitation and its link to relationship risks, not enough is known about cohabitation and its relation to dating and marriage. There is much discussion on whether cohabitation is primarily a precursor to marriage, similar to marriage, or distinct from marriage, or similar to or distinct from being single (Stafford, Kline, & Rankin, 2004, p. 232). On one hand, similar to marriage, cohabitation involves established patterns of behavior, shared living quarters, and, for a substantial number of couples, having a child (Brown & Booth, 1996). On the other hand, cohabiting couples are seen as having less commitment and greater individual autonomy (Forste & Tanfer, 1996; Newcomb, 1986; Nock, 1995). In the United States, cohabiting couples also do not share the same legal status as married couples. Indeed, currently cohabiting couples are more likely to end their relationships (before marriage) than are married couples. Additionally, cohabiting couples who do go on to marry are more likely to divorce than those who did not cohabit before marriage (Bumpass & Sweet, 1989; Kamp Dush, Cohan, & Amato, 2003; Lillard,

Annie C. Hsueh, Kristen Rahbar Morrison, and Brian D. Doss, Department of Psychology, Texas A&M University. Correspondence concerning this article should be addressed to Annie C. Hsueh, Department of Psychology, Texas A&M University, 4235 TAMU, College Station, TX 77843-4235. E-mail: ahsueh@tamu.edu

Brien, & Waite, 1995; Teachman & Polonko, 1990). In contrast to the growing literature on the differences and similarities between cohabiting and married couples, there are virtually no studies comparing dating and cohabiting couples. The literature to date on how dating, cohabiting, and married individuals differ has focused on a limited number of areas such as disagreement, intimate partner violence, mental health status, parenting, sex, and commitment. Furthermore, the existing literature has primarily focused on the comparison between cohabitation and marriage rather than including a comparison of cohabitation and dating. We review this literature below and then suggest important ways in which it can be expanded. Research has yielded mixed ndings on whether cohabiting couples have higher rates of disagreement than do married couples (Brown & Booth, 1996; Nock, 1995; Skinner, Bahr, Crane, & Call, 2002; Stafford et al., 2004), with most studies indicating that there is no difference between the two groups rates of disagreement after a number of variables are controlled. Specically, after controlling for various demographic and relationship variables, there were no signicant differences between cohabiting and married couples in the frequency of disagreements in the past year in the areas of household tasks, money, time spent together, sex, whether to have children, and in-laws (Nock, 1995); the number of couples who argue heatedly (Stafford et al., 2004); and the perceived levels of open disagreements (Skinner et al., 2002). In contrast, a fourth study that also controlled for demographic and relationship variables found that cohabiting individuals experienced a greater overall frequency of disagreement than married individuals in the
236

REPORTS OF PROBLEMS IN COHABITING RELATIONSHIPS

237

areas of money, time together, household tasks, sex, and planning for a child (Brown & Booth, 1996). However, comparison among studies is difcult because the studies used different denitions of disagreements and arguments. Intimate partner violence is more common and more severe in cohabiting couples than in both dating and married couples (e.g.,Brownridge & Halli, 2000; Kline et al., 2004; Stets & Straus, 1989). As compared with married women, cohabiting women were approximately three times more likely to report being the victim of domestic violence and twice as likely to report being the perpetrator of that violence (Kessler, Molnar, Feurer, & Appelbaum, 2001). After controlling for age, education, and occupation, cohabiting couples still had the highest assault rate, followed by dating and then married couples (Stets & Straus, 1989). However, when differences in demographic and social factors (social support and social control) were accounted for, cohabiting and married couples reported similar rates of aggression (Stets, 1991). This result suggests that cohabiting couples higher rates of aggression may be partially spurious and/or indirect (Stets, 1991). Moreover, most (Brown, 2000; Brown, Bulanda, & Lee, 2005; Marcussen, 2005; Stafford et al., 2004), but not all (Horwitz & White, 1998), studies have suggested that cohabiting individuals have more depressive symptoms than do married individuals. Brown et al. (2005) showed that cohabiting individuals had more depressive symptoms than married individuals after controlling for economic resources, social support, and physical health. Additionally, Murcussen (2005) showed that remaining differences in depression between cohabiting and married individuals could be explained by differences in coping resources and relationship quality. In addition to depressive symptoms, cohabiting individuals, especially men, experience more problems with alcohol than do married and single individuals (Horwitz & White, 1998; Marcussen, 2005). These differences in alcohol problems persisted even after controlling for prior levels of alcohol problems, unconventionality, relationship characteristics, and demographic characteristics (Horwitz & White, 1998) and for socioeconomic factors, social resources, relationship commitment, and relationship stability (Marcussen, 2005). Another study (Kamp Dush & Amato, 2005) also showed that married individuals reported the highest level of subjective wellbeing, followed by cohabiting individuals, then by dating individuals. Combined, these results suggest that cohabiters may experience more problems with mental health issues than do married individuals. There is limited research examining how parenting differs across married and cohabiting families. A recent study that examined the effect of family structure and parenting in low-income families (Gibson-Davis, 2008) found that cohabiting and married mothers and fathers were statistically indistinguishable on positive engagement, spanking, and instrumental support. Paternal parenting outcomes were equivalent for cohabiting and married fathers. However, paternal parenting behaviors were affected when the father maintains a separate household. Examinations of sexual frequency differences between cohabiting and married individuals revealed that cohabiting

individuals reported somewhat more frequent sex than married individuals, but the difference in frequency of sex was not signicant. However, married men but not married women were more likely than cohabiting and dating individuals to report that they were extremely satised with the physical nature of their sexual relationship (Waite, 1995). Moreover, both married men and women were more likely than cohabiting and dating individuals to report that they were extremely satised with the emotional aspects of their sex lives (Waite, 1995; Waite & Joyner, 2001). Finally, cohabiting couples have been found to report less commitment to their relationship and greater individual autonomy (e.g., Forste & Tanfer, 1996; Newcomb, 1986; Nock, 1995). Indeed, a large-scale random sample of engaged, married, and cohabiting individuals demonstrated that premarital and nonmarital cohabitation were associated with lower levels of interpersonal commitment (Stanley et al., 2004). Additionally, cohabiting individuals were less sexually exclusive with their partner than were married individuals and more similar to dating individuals in their sexual exclusivity (Forste & Tanfer). Unfortunately, the vast number of research studies on cohabitation have involved large quantitative surveys that, although having the advantage of asking questions of many different types of people, place important constraints on the richness of questions that can be asked. In particular, questions in large-scale surveys are typically closed-ended (often yesno or Likert-type scale questions) and focused on narrow issues determined before the survey is conducted. This reliance on quantitative questionnaires is problematic for three reasons. First, the literature on cohabitation is relatively new and undeveloped; therefore, qualitative research is necessary to rst gather needed information to develop subsequent theories and structured questionnaires. Although the previous quantitative surveys have been excellent for testing a priori hypotheses about behaviors and attitudes or moderators of constructs of interest, they do not generally provide qualitative data from which new, broader theories can be generated. Second, compounding the limited knowledge base on cohabiting couples, most studies of cohabitation have used extremely short measures (e.g., three to ve items) to assess relationship problems, which likely do not adequately tap the universe of relationship problems. As such, these studies are unlikely to reveal problems or group differences that were not hypothesized. Open-ended reports, however, allow participants to report on an innite number of possible relationship problems and are not restricted by a limited theoretical or empirical knowledge base. Notably, with the exception of commitment, the literature to date has focused more on the presence of negative aspects of relationships (e.g., how often arguments occur) rather than the absence of positive aspects of relationships (e.g., how often affection is displayed). In contrast, in qualitative studies, participants were free to describe both the presence of negative aspects and the absence of positive aspects in their relationship. Finally, the use of structured questionnaires may serve to suggest problems that people would not otherwise have reported, possibly creating the appearance of important group differences when none exist.

238

HSUEH, MORRISON, AND DOSS

An exploratory, qualitative study will expand the breadth of knowledge of relationship types and inform future theory and research. To our knowledge, only one study has asked cohabiting and married couples open-ended questions about problems in their relationship (Cunningham, Braiker, & Kelley, 1982). In their small sample of young, childless couples, the reported relationship problems were generally similar for the two groups. However, given that this sample was collected before the early 1980s and the rates and function of cohabitation have changed signicantly since that time (Bumpass & Lu, 2000; Smock, 2000), it is unclear whether these results would generalize to couples in current cohabiting relationships. To address these limitations in the existing literature, we had three major aims in this study. First, we examined the types of relationship problems that cohabiting individuals report as being most problematic in their relationships in response to an open-ended question about relationship concerns. Second, we examined whether there are differences between dating, cohabiting, and married individuals in the reported frequency of various problems encountered in their relationships. Finally, we determined whether demographic (income, education, and age) and relationship factors (length of relationship, relationship satisfaction, and presence or absence of children) explained the relationship status differences in reports of broad and specic relationship problems.

Rogge, 2007). Married participants relationship adjustment (M 20.67, SD 5.91) was signicantly lower than cohabiting (M 22.5, SD 5.71), t(1027) 4.16, p .001, and dating participants (M 23.31, SD 5.24), t(386.89) 6.43, p .001, relationship adjustment; cohabiting and dating participants did not differ signicantly from each other. In addition to the participants described above, we excluded a total of 242 individuals who provided information on their top relationship problems from this study. Participants responses were excluded if a participant was younger than 18 (n 16), not currently in a relationship (n 39), did not provide data about relationship status (n 41), did not provide data about the his or her own or partners gender (n 21), did not provide length of relationship (n 44), or provided erroneous data (n 1).1 Additionally, the Web site collected optional information that allowed for identication of partners in the same relationship. When both partners completed the Web site survey, only one partners data were randomly selected for inclusion, which led to the exclusion of 415 participants. Finally, 112 individuals in homosexual relationships were excluded because we felt that the nature of cohabitation and marriage for individuals in homosexual relationships was likely different than for individuals in heterosexual relationships.

Procedure
Data collection. After providing informed consent, individuals in this study participated anonymously in a larger study taking approximately 20-30 min that examined the frequency and acceptance of various partner behaviors (Doss & Christensen, 2006). Participants were recruited as part of national (e.g., Newsweek, New York Times Book Review, National Public Radio interview) and local (e.g., local newspaper) media coverage of a self-help book designed to improve ones relationship. Participants in the larger study were given the option of completing an additional part of the survey, in which they typed their response to the item In a sentence or two, please describe the area of your relationship that causes the most problems for you personally (not what your partner thinks is the biggest problem). Only the participants who answered this question (22.5% of the participants from the larger study) were included in the current study. Participants who chose to answer the optional question, as compared with those who did not, were less likely to be married (60% vs. 66%), female (64% vs. 75%), and Caucasian (77% vs. 83%). However, when mean differences in age, relationship length, number of children, household income, education, and relationship satisfaction were examined, only the difference in relationship satisfaction met Cohens (1988) criteria for at least a small effect, with those who answered the
1 The total number of participants excluded for various reasons is greater than the overall number of excluded participants because there was overlap in the reasons for exclusion. For example, 1 participant may have been excluded from the study both for being younger than 18 and for leaving out information about relationship length.

Method Participants
Participants responded to an Internet survey about their current relationship. The participants in this study were 801 married individuals (512 women and 289 men), 231 individuals in cohabiting relationships (157 women and 74 men), and 220 individuals in dating relationships (151 women and 69 men). They ranged in age from 18 to 72 years, with a mean of 35.95 years (SD 11.25). The mean relationship length was 12.18 years for married participants (SD 9.16), 3.48 years for cohabiting participants (SD 2.85), and 2.16 for dating participants (SD 2.13). Many of the married participants (60.9%) had children with their current partner (mean number of children 1.19, SD 1.19), whereas 20.3% of cohabiting participants and 11.4% of dating participants had children with their current partner (cohabiting participants, mean number of children 0.15, SD 0.44; dating participants, mean number of children 0.04, SD 0.27). Among all participants, 78.8% identied themselves as White, 6.5% as Hispanic, 5.2% as African American, 3.9% as Asian American, 1.0% as Native American, 0.8% as Middle Eastern, 1.5% as biracial, and 2.3% as other. Participants had a mean of 15.54 years of education (SD 2.90), and the median household income bracket was $60,000 $79,999 per year. Participants completed the seven-item Dyadic Adjustment Scale (Hunsley, Best, Lefebvre, & Vito, 2001), on which scores below 21.5 represent relationship adjustment in the distressed range (i.e., a full Dyadic Adjustment Scale score below 98; Funk &

REPORTS OF PROBLEMS IN COHABITING RELATIONSHIPS

239

optional question being somewhat more satised with their relationships (d 0.22). There was no limit to how many problems participants could report. Overall, participants reported an average of 1.62 problems, with cohabiting individuals (M 1.59, SD 0.91) reporting similar numbers of relationship problems as dating (M 1.49, SD 0.83) and married (M 1.67, SD 0.95) individuals. However, dating participants reported signicantly fewer relationship problems than did married couples, t(391.37) 2.71, p .01, which is not surprising given that dating individuals were signicantly more satised in their relationships. As compensation for their participation, individuals were provided with feedback comparing their relationship acceptance to normative data; no feedback was provided on the basis of the qualitative responses analyzed in this study. For couples reporting physical abuse, the site also provided contact information for national domestic violence hotlines. All procedures in this study were approved by the University of California, Los Angeles, Institutional Review Board.

Measures
As part of a larger assessment battery, the following questionnaires were administered. Demographics. All participants were asked for the following demographic information: age, participants gender and partners gender, relationship duration, number of children with current partner, ethnicity, years of education, and annual household income. Additionally, participants were asked to categorize their current relationship status as one of the following: married, living together, dating, divorced or legally separated, single, or other. Relationship satisfaction. In the current study, global relationship satisfaction was measured with a single item from the Dyadic Adjustment Scale (Spanier, 1976), because the other items on the scale (e.g., disagreement and sexual frequency) could potentially overlap with participants reported relationship problem. Specically, we used Item 31 from the Dyadic Adjustment Scale (Spanier, 1976); this item asks participants to describe the degree of happiness, all things considered, of your relationship on a 7-point Likert scale ranging from 0 (extremely unhappy) to 6 (perfect) (Spanier, 1976). This single item of satisfaction has been found to provide high levels of information about individuals relationship satisfaction across a broad range of relationship functioning (Funk & Rogge, 2007). Coding. Participants open-ended responses were coded using a modied version of a coding system previously used to code problems reported by couples seeking marital therapy (Doss, Simpson, & Christensen, 2004). Before nal coding began for this study, we reviewed common, published, detailed quantitative measures of relationship problem and strength areas such as the Areas of Change Questionnaire (Weiss & Birchler, 1975), the Spouse Observation Checklist (Weiss & Perry, 1979), and the Frequency and Acceptability of Partner Behavior Inventory (Doss & Christensen, 2006). When important areas were not represented in the existing coding system, they were added for this study. After the a priori relationship areas were

added, the coding system was further rened during the initial training of coders. During this process, codes were added to capture common unanticipated statements of problems, to reduce ambiguity in codes, or both. At the same time, some codes that were infrequently seen in the written responses and were deemed theoretically less interesting were combined with other codes or dropped from the coding system. The nal coding system used in this study included 65 specic problem codes that were grouped a priori into 16 broad relationship problem codes (Table 1). Three undergraduate coders were trained thoroughly in the coding system by Kristen Rahbar Morrison before they began coding participants statements. Coders were unaware of the studys hypotheses and the participants relationship status; furthermore, to minimize any bias because of gender or relationship status, all gender-based or relationship status based labels (i.e., he/she, husband/wife, and boyfriend/ girlfriend) were replaced with gender-neutral and statusneutral words (i.e., they or partner) before coding. Throughout the entire coding process, whenever two coders agreed on a code for a certain problem, that code was recorded as the nal code. When all three coders disagreed on how to code a certain problem, the coders and Kristen Rahbar Morrison discussed the discrepancy during a weekly coding meeting and came to a consensus about the best code for the reported problem. Reliability of the coding system was computed by examining agreement between the three coders in a random subset of 300 responses (approximately 24% of the total sample) using codes independently generated by each coder before the weekly meeting. The kappa reliability statistics for this random subset were 0.84 for the broad problem codes and 0.75 for the specic codes. These reliability statistics are in the almost perfect and substantial range, respectively (Landis & Koch, 1977).

Results Cohabiting Individuals Top Relationship Problems

We conducted descriptive analyses to examine cohabiting individuals top relationship concerns. Table 1 presents the percentage of cohabiting individuals who endorsed each of the 16 broad codes, as well as any specic codes that were endorsed by at least 5% of the cohabiting individuals. We calculated the percentage of individuals who reported child parenting difculties on the basis of the subset of participants who had children. The top ve broad relationship concerns considered most problematic by cohabiting individuals were problems in specic areas of their current relationship (reported by 29% of cohabiting individuals; e.g., We are living in a hotel due to a temporary transfer through my job. We are unsettled and get a bit stir crazy at times); individual problems, such as own or partners personality or problematic habits (19.5%; e.g., My partner is pessimistic toward the future); general communication (15.2%; e.g, We differ in the area of communication on how much is enough); arguments (15.2%; e.g., We cannot resolve our problems. We cant even discuss them like

240

HSUEH, MORRISON, AND DOSS

Table 1 Percentage of Dating, Cohabiting, and Married Individuals Who Reported Various Broad and Specic Relationship Problems
Reported problem Specic areas of current relationship Relationship stress/unmet needs specic to couple Issues with nances Individual problems Personality or moods Health problems and other individual problematic habits, patterns, interests, or afliation General communication Unspecied lack of communication/disagreement about how to communicate/partner does not listen to me/avoidance of conict Arguments General tension/anger Cannot resolve conicts Emotional affection/distance Social activities/time together Desire to have more time/better quality time with partner Other issues of time together/social activities or interests Lack of physical affection/sex Frequency or quality of sexual activity Trust/jealousy Criticalness Relationship commitment/security Substance use/abuse Child/parentinga Problems with previous relationships Mental health disorder Indelity/irting Abuse/violence % within dating individuals 24.1 12.7 5.0 12.7 4.1 8.2 18.2 8.2 5.9 4.5 0.5 12.7 16.8 7.3 6.8 7.7 2.7 8.6 7.3 13.2 2.3 0 3.2 0.5 2.7 0.5 % within cohabiting individuals 29.0 13.9 8.7 19.5 10.8 10.0 15.2 10.4 15.2 8.7 6.1 13.4 10.8 5.6 5.2 10.0 6.9 6.9 6.5 5.2 4.8 4.0 3.5 1.7 1.7 0.4 % within married individuals 30.3 10.5 12.0 16.1 6.7 8.0 16.1 11.2 7.7 5.4 2.2 19.7 12.1 6.7 5.4 21.2 14.7 4.6 9.5 1.4 2.6 4.8 0.9 1.2 2.2 1.4

Note. All broad codes are presented; only the specic problems that were endorsed by at least 5% of the cohabiting individuals are reported here. a The percentages presented for child/parenting difculties were calculated using only the subsets of participants in each relationship type who had children. Twenty-three married individuals, 1 cohabiting individual, and no dating individuals reported child/parenting difculties.

adults); and emotional affection or distance (13.4%; e.g., We dont tell each other that we love each other on a regular basis). In contrast, cohabiting individuals were least likely to report problems with abuse or violence (0.4%; e.g., being verbally abusive when drinking), indelity or irting (1.7%; e.g., I get angry with him mostly for looking at pretty women and letting them irt with him), a mental health disorder in themselves or their partner (1.7%; e.g., We both suffer from depression), problems stemming from previous relationships (3.5%; e.g., Im not divorced yet), or child or parenting difculties (4.0%; e.g., Not following through with disciplining the kids).

cantly differed by relationship type was split into its individual problem codes and again tested for differences by relationship type.3 To conduct these analyses, participants relationship types dating, cohabiting, or marriedwere dummy coded with cohabiting individuals as the comparison group. We then conducted hierarchical logistic regressions to test differences in how dating, cohabiting, and married individuals differed in
Given this articles focus on cohabiting individuals, we report only how cohabiting individuals compare with dating or married individuals. However, in all cases except arguments, comparisons that were signicant between cohabiting and married individuals were also signicant when examining differences between dating and married individuals. Except for arguments, cohabiting individuals reported an intermediate frequency of problems. Full results are available from Annie C. Hsueh. 3 Given the heterogeneous nature of the specic problems under the global code of specic areas of current relationship, we analyzed all six specic problems within this broad area separately.
2

Analyses to Test for Group Differences

We conducted analyses in three steps. First, after controlling for the gender of the respondent to control for the possibility that men and women would report problems at different frequencies, we examined the 16 broad relationship problem codes to determine whether they differed by relationship type.2 Second, any broad code that signi-

REPORTS OF PROBLEMS IN COHABITING RELATIONSHIPS

241

their reported problems; gender was entered in Step 1, the dummy variables for relationship type were entered in Step 2, and the two two-way interactions between gender and relationship type were entered in Step 3. For the child or parenting problem code, we included only those who had children in the analyses. Finally, in the third set of analyses, when the broad or specic problem codes were found to differ by relationship type, we explored whether demographic and relationship factors (other than relationship type) could explain the differences by relationship type. Specically, gender was entered in Step 1, demographic factors (income, education, and age) were entered in Step 2, and relationship factors (relationship duration, global relationship satisfaction, and children with the current partner) were entered together in Step 3. After accounting for these variables, relationship type (Step 4) and Gender Relationship Type interactions (Step 5) were subsequently examined to determine whether they remained signicant predictors of reported problems.

Dating Versus Cohabiting

Broad problems. Comparisons of dating and cohabiting individuals reported frequency of all 16 broad problems are presented in Table 2. The reported frequency of various broad problems encountered in dating and cohabiting relationships were similar, with two exceptions. First, dating individuals were less than half as likely as cohabiting individuals to report problems with arguments (p .01, odds ratio [OR] 0.35). Second, dating individuals were more than twice as likely as cohabiting individuals to report problems with relationship commitment or security (p .01, OR 2.77). There were no signicant Gender Relationship Type interactions. Specic problems. Within the broad problem of arguments, dating individuals were signicantly less likely than cohabiting individuals to report an inability to resolve conicts (Table 2; p .05, OR 0.07). Other specic codes under the broad argument code that did not signicantly

Table 2 Difference in Reported Frequency of All Global Problems and Signicant Specic Problems by Individuals in Dating Versus Cohabiting Relationship and Married Versus Cohabiting Relationship
Comparisons with cohabiting individuals Dating Reported problem General communication Arguments Cannot resolve conicts Criticalness Abuse/violence Emotional affection/distance Lack of emotional intimacy/affection Lack of physical affection/sex Frequency or quality of sexual activity Trust/jealousy Relationship commitment/security Lack of commitment Insecure about partners feelings Individual problems Mental health disorder Substance use/abuse Child/parenting Social activities/time together Indelity/irting Problems with previous relationships Damaged/hurt by previous relationship Other ex/previous relationship issue Specic areas of current relationship Household chores, division of labor Issues with nances Religious/faith differencesa In-laws/extended family/parents Problems with goals for future/values Relationship stress/unmet needs specic to couple b 0.22 1.05 2.65 0.12 0.05 0.06 0.28 0.24 1.02 1.27 0.51 1.35 0.77 17.9 0.51 0.46 0.09 0.25 0.96 0.59 0.75 0.05 0.10 Odds ratio 1.24 0.35 0.07 1.13 1.05 0.94 0.76 1.27 2.77 3.54 0.60 0.26 0.46 .00 1.67 1.59 0.92 0.78 0.39 0.55 2.11 1.05 0.91 b 0.10 0.75 1.05 0.40 1.17 0.48 0.84 0.88 0.83 0.43 1.35 1.94 0.23 0.33 0.59 0.19 0.13 0.30 1.42 2.36 1.49 0.07 0.60 0.38 0.49 0.78 0.32 Married Odds ratio 1.10 0.47 0.35 1.49 3.23 1.61 2.33 2.40 2.29 0.65 0.26 0.14 0.79 0.72 0.55 1.20 1.14 1.35 0.24 0.10 0.23 1.07 1.83 1.46 1.63 0.46 0.73

Note. Dashes indicate the analyses were not conducted because the broad code was not signicantly different between the two groups. a No cohabiting individuals endorsed this code. Therefore, analysis was excluded. p .05. p .01.

242

HSUEH, MORRISON, AND DOSS

differ were frequency or amount of arguments and general tension. Second, within the broad code of relationship commitment and security, dating individuals were signicantly more likely than cohabiting individuals to report problems resulting from a lack of commitment (p .01, OR 3.54); there were no signicant differences between dating and cohabiting participants reports of insecurity about their partners feelings or fear of rejection from their partner. Role of relationship and demographic factors. Both demographic characteristics, 2(3, N 1130) 17.05, p .01, and relationship characteristics, 2(3, N 1130) 15.0, p .01, signicantly predicted differences in reported problems with relationship commitment or security. Relationship characteristics, 2(3, N 1130) 8.49, p .05, predicted the signicant difference in reports of lack of commitment. After controlling for all six relationship and demographic characteristics,4 most of the differences between dating and cohabiting individuals remained signicant. Specically, the difference between dating and cohabiting individuals in the likelihood of reporting problems with arguments (p .1, OR 0.30), relationship commitment and security (p .01, OR 3.34), and lack of commitment (p .01, OR 9.68) remained signicant, whereas the likelihood of reporting problems with inability to resolve conicts was no longer signicant.

problems in lack of physical affection or sex, 2(3, N 1130) 10.59, p .05; frequency or quality of sexual activity, 2(3, N 1130) 11.33, p .05), relationship commitment or security, 2(3, N 1130) 17.05, p .01; insecurity about partners feelings, 2(3, N 1130) 11.80, p .01; and problems with goals for future or values, 2(3, N 1130) 9.56, p .05. Relationship characteristics predicted the signicant difference in reporting problems for all broad and specic codes other than problems with arguments, inability to resolve conicts, and other ex or previous relationship issues. When all six relationship and demographic characteristics were entered together into Step 2, only half of the differences in the reports of broad problems between married and cohabiting individuals remained signicant. Specically, the likelihood of reporting the broad problems of arguments (p .05, OR 0.50) and lack of physical affection or sex (p .05, OR 1.91) remained signicant. However, the broad problems of emotional affection or distance, relationship commitment or security, and problems with previous relationships, along with all specic problems, were no longer signicant.

Discussion
This study advances the understanding of cohabiting individuals relationship concerns and how those concerns differ from those of dating and married individuals. By using participants open-ended responses about the top area of concern in their relationship and a detailed and reliable coding system, this study permitted the examination of relationship concerns without imposing a preexisting framework on such concerns. Given the relative infancy of research exploring the nature of cohabitation, this qualitative approach was necessary to avoid omitting or overemphasizing differences. Additionally, in this study we explored demographic and relationship differences as possible explanations of group differences and determined whether group differences existed after controlling for these predictors. The results of this study suggest several avenues for future research.

Married Versus Cohabiting

Broad problems. Comparisons of married and cohabiting individuals reported frequency of all 16 broad problems are presented in Table 2. Cohabiting and married individuals differed in their reported frequency of 6 of the 16 broad problems. On one hand, married individuals were much less likely than cohabiting individuals to report problems with arguments (p .01, OR 0.47), relationship commitment and security (p .01, OR 0.26), and problems with a previous relationship (p .01, OR 0.24). On the other hand, married individuals were signicantly more likely than cohabiting individuals to report problems with emotional affection and distance (p .05, OR 1.61) and lack of physical affection or sex (p .01, OR 2.40). There were no signicant Gender Relationship Type interactions. Specic problems. Married individuals were signicantly less likely than cohabiting individuals to report the following problems: an inability to resolve conicts (Table 2; p .01, OR 0.35), insecurity about their partners feelings (p .01, OR 0.14), being damaged or hurt by a previous relationship (p .05, OR .10), another ex or previous relationship issue (p .05, OR 0.23), and problems with goals for the future or values (p .05, OR 0.46). In contrast, married individuals were signicantly more likely than cohabiting individuals to report lack of emotional intimacy or affection (p .05, OR 2.33) and frequency or quality of sexual activity (p .01, OR 2.29). Role of relationship and demographic factors. Demographic characteristics predicted the difference in reporting

Relationship Areas of Concern for Cohabiting Individuals

This study helped to illuminate the top relationship problems considered by cohabiting individuals to be the most problematic. As reviewed in the introduction, previous research on cohabitation has focused, among other areas, on levels of violence, mental health symptoms, and child difculties. However, using qualitative reports of individuals top relationship problems, this study suggested that cohabiting individuals only infrequently report domestic violence (0.4%), mental health problems (1.7%), or child or parenting difculties (4.0% of individuals with children) as major problems in their relationship.
4 The information on how each relationship and demographic factor individually predicted the likelihood of reporting problems is available from Annie C. Hsueh.

REPORTS OF PROBLEMS IN COHABITING RELATIONSHIPS

243

Instead, the broad relationship areas that were viewed as most problematic for cohabiting individuals were problems in specic areas of their current relationship (especially nances and stress or unmet needs), individual problems (e.g., personality or health problems), general communication, arguments, and emotional affection or distance. Thus, future studies on cohabitation may benet from focusing on these problematic relationship areas. Additionally, given that cohabiting couples tend to seek premarital education at lower rates (Doss, Rhoades, Stanley, Markman, & Johnson, in press; Halford, ODonnell, Lizzio, & Wilson, 2006), it would be important for service providers to advertise those interventions as targeting relationship problems that cohabiting individuals identify frequently in their open-ended reports. Previous research with married couples has suggested many overlaps between the top relationship problems of cohabiting couples revealed in this study. For example, previous research has demonstrated that, like cohabiting couples in this study, couples commonly argue about money (Stanley, Markman, & Whitton, 2002; Storaasli & Markman, 1990) and communication (Storaasli & Markman, 1990) in their rst marriages. However, given that these studies used structured questionnaires and different methodologies, it was important to directly compare cohabiting individuals reported problems with those problems reported by married and dating individuals in this study.

Comparisons of Relationship Problems With Married Individuals

Our results revealed a number of differences in the frequency of reported relationship problems between cohabiting and married individuals. Cohabiting individuals were more likely than married individuals to report problems with arguments, an inability to resolve conicts, poor relationship commitment or security, insecurity about their partners feelings, problems with a previous relationship, being damaged or hurt by a previous relationship, difculties stemming from previous relationships, and disagreement about values and goals for the future. Together, these differences indicate that cohabiting relationships tend to be experienced as more volatile than marriages in that cohabiting individuals tend to argue more, nd their relationships more unstable or insecure, and have more issues with past relationships and with future goals and values. However, we should note that the impact of this volatility on future relationship functioning cannot be determined from these results. Indeed, it may be that couples who experience this volatility may grow stronger as a result of these problems (e.g., Kelley & Thibaut, 1978) or are less likely to slide into marriage (e.g., Stanley, Rhoades, & Markman, 2006). Previous studies (Waite, 1995; Waite & Joyner, 2001) have found that married individuals are more likely than cohabiting individuals to report being extremely satised with the physical and emotional aspects of their sexual relationships. However, to our knowledge, no studies have examined differences in problematic levels of physical and emotional satisfaction with sex that were the focus of this

study. It is possible that there is more variability in married individuals sexual satisfaction than in cohabiting individuals sexual satisfaction, leaving open the possibility that married individuals would report both more extreme satisfaction and major problems with sex than would cohabiting individuals. Indeed, these results revealed that cohabiting individuals were less likely than married individuals to report problems with emotional affection or distance and a lack of physical affection or sex. To our knowledge, these more favorable results for cohabiting relationships have not been highlighted in previous research. There are at least two potential explanations for these differences. It could be that cohabiting individuals are more likely than married individuals to attend to the emotional and physical aspects of their relationship, leading to fewer problems in these areas. Alternatively, the fewer problems with emotional and physical intimacy in cohabiting individuals may be a reection of the differential separation rates between cohabiting and married individuals; cohabiting individuals with serious problems in the emotional or physical aspects of their relationship may be more likely to terminate their relationship and thus not be included in our data set. Finally, several relationship problems (e.g., domestic violence, mental health problems, and child difculties) that have been found to differ between cohabiting and married individuals were not reported at different rates in this study. This lack of signicance may have resulted because individuals do not consider these issues as major problems in their relationships; however, when asked about these areas on a standardized questionnaire (as in previous research), cohabiting individuals may report higher levels of these problems than do dating or married individuals.

Comparisons With Dating Individuals

Results of this study revealed that dating and cohabiting individuals were mostly similar in their reports of relationship concerns. Indeed, the groups differed on only 2 out of 16 broad codes and 2 out of the 5 specic codes. Specically, cohabiting individuals were more likely to report broad problems with arguments and less likely to report broad problems with relationship commitment or security. Furthermore, these differences remained signicant after accounting for a variety of relationship and demographic variables. As for specic problems, after accounting for all six relationship and demographic variables, cohabiting individuals were more likely to report not being able to resolve conicts and less likely to report problems with lack of commitment than were dating individuals. Therefore, in comparison to dating relationships, cohabiting individuals reported more difculty resolving conicts but also greater relationship stability. Although increased arguments in cohabiting relationships have been documented in previous research (e.g., Brown & Booth, 1996), to our knowledge increased levels of relationship commitment in cohabiting individuals (relative to dating individuals) has not. This nding is especially notable given that after controlling for de-

244

HSUEH, MORRISON, AND DOSS

mographic and relationship factors, cohabiting and married individuals did not report signicantly different rates of problems with commitment. Therefore, it may be that once couples decide to either cohabit or marry, relationship problems around commitment that characterized dating relationships decrease. If supported by future research, these ndings would suggest that researchers should focus on the commitment process that occurs when couples decide to cohabit, not just when cohabiting or dating couples decide to marry.

study cannot explain the difference between married and cohabiting individuals in these areas. Therefore, these results seem to suggest that both selection and experience explanations may explain the frequency of reported relationship problems; however, given the limitation of cross-sectional studies to address this question, longitudinal research will ultimately be needed.

Limitations and Implications

In considering this studys results, it is important to acknowledge its limitations. First, the data gathered focused exclusively on participants current relationship status and problems. Therefore, we were unable to examine whether current problems in the relationship were inuenced by the participants relationship history. For example, within married individuals, we were unable to examine whether there was a difference in reported problems between participants who did and did not cohabit before marriage. Additionally, we did not collect information on whether cohabiting individuals had plans to get married. Previous research has suggested that cohabitation before and after engagement is an important distinction, with couples who cohabit before engagement likely to slide into marriage (i.e., the inertia theory; Stanley et al., 2006). Second, participants in this study were a convenience sample collected over the Internet who further self-selected from the larger study by taking the time to complete an optional section on the top problems in their relationship. Thus, both the convenience and the self-selection nature of this sample likely limits the studys generalizability. Indeed, in comparison to the general U.S. population, our sample was more predominantly White, was more educated, and had a higher income. Despite these limitations, this study advances knowledge related to the relationship concerns of individuals in dating, cohabiting, and married relationships. In particular, this study documents common relationship concerns of cohabiting individuals and highlights both strengths and weaknesses of these relationships in comparison to dating and marital relationships. On the basis of the results from these open-ended responses, it may be possible to develop structured questionnaires that would be sensitive to both the most common problems in cohabiting relationships and the differences between cohabiting and other types of relationships. Additionally, such knowledge can assist in the development and renement of relationship interventions to make them more applicable to cohabiting individuals. For example, given that cohabiting individuals report more problems with arguments but fewer problems with emotional distance or physical affection as compared with married individuals, communication skills designed to facilitate problem solving rather than emotional expression may be more appropriate for the average cohabiting couple. In addition, skills training in the areas of communication, and especially conict resolution, may be more helpful than efforts to strengthen the couples emotional connection,

The Cohabitation Effect: Selection or Experience?

The association between premarital cohabitation and poorer relationship outcomes has been attributed to two fundamental explanations: selection and experience (e.g., Brown & Booth, 1996; Smock, 2000; Stanley et al., 2006). The selection explanation suggests that people who do and do not choose to cohabit before marriage differ in certain characteristics (such as income, education, age, and religiosity) and that the specic characteristics may contribute to the poorer relationship functioning observed in cohabiting couples. Indeed, research has shown that individuals who cohabit before marriage tend to be of slightly lower socioeconomic status and slightly more liberal (for a review, see Smock, 2000). In contrast, the experience explanation suggests that the act of living together before marriage may increase risk for relationship distress. For example, researchers have hypothesized that couples who cohabit before marriage (and especially before engagement) may be more likely to marry simply because the barriers to ending the relationship increase as a result of cohabitation (e.g., shared mortgage, pets, and furniture; Stanley et al., 2006). The selection and experience explanations are not mutually exclusive, and both have received empirical support (see Smock, 2000). Although in this study we investigated only a limited number of demographic variables, it was notable that the group of demographic variables (which can be attributed to a selection explanation) was a signicant predictor of the frequency of two broad problems and three specic problems. However, even after controlling for these demographic characteristics, relationship characteristics signicantly predicted the frequency of four broad problems and six specic problems. Unlike demographic variables, however, relationship characteristics likely represent a mixture of selection and experience effects. For example, less satised couples may choose to cohabit rather than get married, and the experience of cohabitation may make them even less satised. Finally, even after accounting for all six relationship and demographic variables, the reported frequency of two relationship problems remained signicantly different between married and cohabiting individuals. Specically, cohabiting individuals were more likely than married individuals to report problems with arguments but were less likely to report problems with a lack of physical affection and sex. Thus, the selection variables (either demographic or relationship factors) examined in this

REPORTS OF PROBLEMS IN COHABITING RELATIONSHIPS

245

which may already be satisfactory. In sum, although important research on cohabitation remains, this study lays a foundation for understanding the common relationship problems in cohabiting relationships and highlights similarities and differences among cohabiting, dating, and marital relationships.

References
Brown, S. L. (2000). The effect of union type on psychological well-being: Depression among cohabitors versus marrieds. Journal of Health and Social Behavior, 41, 241255. Brown, S. L., & Booth, A. (1996). Cohabitation versus marriage: A comparison of relationship quality. Journal of Marriage and the Family, 58, 668 678. Brown, S. L., Bulanda, J. R., & Lee, G. R. (2005). The signicance of nonmarital cohabitation: Marital status and mental health benets among middle-aged and older adults. Journals of Gerontology: Psychological Sciences and Social Sciences, 60B, S21S29. Brownridge, D. A., & Halli, S. S. (2000). Living in sin and sinful living: Toward lling a gap in the explanation of violence against women. Aggression and Violent Behavior, 5, 565583. Bumpass, L., & Lu, H.-H. (2000). Trends in cohabitation and implications for childrens family contexts in the United States. Population Studies, 54, 29 41. Bumpass, L., & Sweet, J. A. (1989). National estimates of cohabitation. Demography, 26, 615 625. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Cunningham, J. D., Braiker, H., & Kelley, H. H. (1982). Maritalstatus and sex differences in problems reported by married and cohabiting couples. Psychology of Women Quarterly, 6, 415 427. Doss, B. D., & Christensen, A. (2006). Acceptance in romantic relationships: The Frequency and Acceptability of Partner Behavior Inventory. Psychological Assessment, 18, 289 302. Doss, B. D., Rhoades, G. K., Stanley, S. M., Markman, H. J., & Johnson, C. A. (in press). Differential use of premarital education in rst and second marriages. Journal of Family Psychology. Doss, B. D., Simpson, L. E., & Christensen, A. (2004). Why do couples seek marital therapy? Professional Psychology: Research and Practice, 35, 608 614. Forste, R., & Tanfer, K. (1996). Sexual exclusivity among dating, cohabiting, and married women. Journal of Marriage and the Family, 58, 33 47. Funk, J. L., & Rogge, R. D. (2007). Testing the ruler with item response theory: Increasing precision of measurement for relationship satisfaction with the couples satisfaction index. Journal of Family Psychology, 21, 572583. Gibson-Davis, C. M. (2008). Family structure effects on maternal and paternal parenting in low-income families. Journal of Marriage and Family, 70, 452 465. Halford, W. K., ODonnell, C., Lizzio, A., & Wilson, K. L. (2006). Do couples at high risk of relationship problems attend premarriage education? Journal of Family Psychology, 20, 160 163. Horwitz, A. V., & White, H. R. (1998). The relationship of cohabitation and mental health: A study of a young adult cohort. Journal of Marriage and the Family, 60, 505514. Hunsley, J., Best, M., Lefebvre, M., & Vito, D. (2001). The seven-item short form of the Dyadic Adjustment Scale: Further evidence for construct validity. American Journal of Family Therapy, 29, 325335. Kamp Dush, C. M., & Amato, P. R. (2005). Consequences of

relationship status and quality for subjective well-being. Journal of Social and Personal Relationships, 22, 607 627. Kamp Dush, C. M., Cohan, C. L., & Amato, P. R. (2003). The relationship between cohabitation and marital quality and stability: Change across cohorts? Journal of Marriage and Family, 65, 539 549. Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence. New York: Wiley. Kessler, R. C., Molnar, B. E., Feurer, I. D., & Appelbaum, M. (2001). Patterns and mental health predictors of domestic violence in the United States: Results from the National Comorbidity Survey. International Journal of Law and Psychiatry, 24, 487508. Kline, G. H., Stanley, S. M., Markman, H. J., Olmos-Gallo, P. A., St. Peters, M., Whitton, S. W., et al. (2004). Timing is everything: Pre-engagement cohabitation and increased risk for poor marital outcomes. Journal of Family Psychology, 18, 311318. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159 174. Lillard, L. A., Brien, M. J., & Waite, L. J. (1995). Premarital cohabitation and subsequent marital dissolution: A matter of self-selection? Demography, 32, 437 457. Marcussen, K. (2005). Explaining differences in mental health between married and cohabiting individuals. Social Psychology Quarterly, 68, 239 257. Newcomb, M. D. (1986). Cohabitation, marriage and divorce among adolescents and young adults. Journal of Social and Personal Relationships, 3, 473 494. Nock, S. L. (1995). A comparison of marriages and cohabiting relationships. Journal of Family Issues, 16, 5376. Skinner, K. B., Bahr, S. J., Crane, D. R., & Call, V. R. A. (2002). Cohabitation, marriage, and remarriage: A comparison of relationship quality over time. Journal of Family Issues, 23, 74 90. Smock, P. J. (2000). Cohabitation in the United States: An appraisal of research themes, ndings, and implications. Annual Review of Sociology, 26, 120. Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 1528. Stafford, L., Kline, S. L., & Rankin, C. T. (2004). Married individuals, cohabiters, and cohabiters who marry: A longitudinal study of relational and individual well-being. Journal of Social and Personal Relationships, 21, 231248. Stanley, S. M., Markman, H. J., & Whitton, S. W. (2002). Communication, conict and commitment: Insights on the foundations of relationship success from a national survey. Family Process, 41, 659 675. Stanley, S. M., Rhoades, G. K., & Markman, H. J. (2006). Sliding versus deciding: Inertia and the premarital cohabitation effect. Family Relations, 55, 499 509. Stanley, S. M., Whitton, S. W., & Markman, H. J. (2004). Maybe I do: Interpersonal commitment and premarital or nonmarital cohabitation. Journal of Family Issues, 25, 496 519. Stets, J. E. (1991). Cohabiting and marital agglession: The role of social isolation. Journal of Marriage and the Family, 53, 669 680. Stets, J. E., & Straus, M. A. (1989). The marriage license as a hitting license: A comparison of assaults in dating, cohabiting, and married couples. Journal of Family Violence, 4, 161180. Storaasli, R. D., & Markman, H. J. (1990). Relationship problems in the early stages of marriage: A longitudinal investigation. Journal of Family Psychology, 4, 80 98. Teachman, J. D., & Polonko, K. A. (1990). Cohabitation and marital stability in the United States. Social Forces, 69, 207220.

246

HSUEH, MORRISON, AND DOSS Weiss, R. L., & Perry, B. A. (1979). Assessment and treatment of marital dysfunction. Eugene, OR: Oregon Marital Studies Program.

Waite, L. J. (1995). Does marriage matter? Demography, 32, 483507. Waite, L. J., & Joyner, K. (2001). Emotional satisfaction and physical pleasure in sexual unions: Time horizon, sexual behavior, and sexual exclusivity. Journal of Marriage and the Family, 63, 247264. Weiss, R. L., & Birchler, G. R. (1975). Areas of Change questionnaire. Unpublished measure, University of Oregon.

Received May 19, 2008 Revision received December 22, 2008 Accepted December 31, 2008

Members of Underrepresented Groups: Reviewers for Journal Manuscripts Wanted

If you are interested in reviewing manuscripts for APA journals, the APA Publications and Communications Board would like to invite your participation. Manuscript reviewers are vital to the publications process. As a reviewer, you would gain valuable experience in publishing. The P&C Board is particularly interested in encouraging members of underrepresented groups to participate more in this process. If you are interested in reviewing manuscripts, please write APA Journals at Reviewers@apa.org. Please note the following important points: To be selected as a reviewer, you must have published articles in peer-reviewed journals. The experience of publishing provides a reviewer with the basis for preparing a thorough, objective review. To be selected, it is critical to be a regular reader of the ve to six empirical journals that are most central to the area or journal for which you would like to review. Current knowledge of recently published research provides a reviewer with the knowledge base to evaluate a new submission within the context of existing research. To select the appropriate reviewers for each manuscript, the editor needs detailed information. Please include with your letter your vita. In the letter, please identify which APA journal(s) you are interested in, and describe your area of expertise. Be as specic as possible. For example, social psychology is not sufcientyou would need to specify social cognition or attitude change as well. Reviewing a manuscript takes time (1 4 hours per manuscript reviewed). If you are selected to review a manuscript, be prepared to invest the necessary time to evaluate the manuscript thoroughly.