Co-Rumination Communication Patterns 1 Co-Rumination Communication Patterns in Friendships and Romantic Relationships

Depression is a disease that affects more women than men. One explanation why women are more frequently diagnosed with depression is that they cope with problems differently than men do. One such coping style, co-rumination, is the process of repeatedly talking about one’s problems without resolving them. While co-rumination is positively related to the quality of the friendship, it also correlates with higher instances of depression (Rose, 2002). Additionally, past research has found an interaction to exist between gender and relationship regarding corumination. Basically, men co-ruminated more with partners whereas women co-ruminated more with friends, though, overall, women significantly co-ruminated more than men did (Assad, 2004; Assad & Haley, 2003). Additionally, previous research has shown that listening behaviors are important factors in determining whether or not the speaker should continue to self-disclose (Reis & Shaver, 1988), and, therefore, may be relevant to co-rumination patterns. The aims of this study were to replicate the results of past findings on co-rumination, to examine gender differences in nonverbal and verbal listening skills, and to examine the degree to which dyad’s co-rumination scores correlated. Participants consisted of 96 undergraduates at a small, liberal arts college. Data was collected through 3 videotaped conversations (neutral, negative, and positive) and through a questionnaire. Among the results, the data partially replicated past findings. Additionally, there was found to be a significant interaction between gender and relationship type regarding listening behavior, F(1, 92) = 8.97, p = .004. Co-Rumination Communication Patterns 3 Co-Rumination Communication Patterns in Friendships and Romantic Relationships After age 13, significantly more girls than boys are diagnosed with depression (NolenHoeksema, 2001). Many theories exist on why depression is consistently found more frequently in women than in men, which include differences in biological hormones, social and economic stresses, and coping styles (Nolen-Hoeksema, 2001). Coping styles may influence depression because they determine if individuals respond to problems productively or destructively. Rumination, or “coping with negative mood that involves self-focused attention” (Treynor, Gonzalez, & Nolen-Hoeksema, 2003) is one type of coping mechanism that has been positively linked to depression (Nolen-Hoeksema, 2001; NolenHoeksema, Parker, & Larson, 1994; Treynor et al., 2003). Nolen-Hoeksema et al. (1994) has suggested that, because rumination is the tendency to dwell on problems without resolving them, it may inhibit the ability to find solutions to these problems, which in turn perpetuates a depressed state. Amanda Rose (2002) found a particular type of rumination, co-rumination, to be more strongly present in female than in male friendships. Co-rumination is the act of “extensively discussing and revisiting problems, speculating about problems, and focusing on negative feelings.” The difference between rumination and co-rumination is that co-rumination is social while rumination is not. While rumination only involves inward reflection on problems, corumination includes outwardly vocalizing these problems to another person. Rose (2002) conducted a study to examine the differences between boys and girls (children in the 3 and 5 grades, and adolescents in the 7 and 9 grades) in the amount of co-rumination each group exhibited in same-sex friendships. Rose (2002) found that girls reported co-ruminating significantly more than boys in both childhood and early adolescence. Furthermore, while boys’ Co-Rumination Communication Patterns 4 co-rumination patterns remained the same with age, adolescent girls co-ruminated significantly more than younger girls. Interestingly, this increase in co-rumination behaviors appeared at the same age in which Nolen-Hoeksema (2001) first found higher levels of depression to start appearing in girls relative to boys. Because of this gender difference, Rose (2002) hypothesized that co-rumination may be a factor leading to depression in women. Additional co-rumination research has been conducted on college populations (Assad, 2004; Assad & Haley, 2003). In addition to extending Rose’s (2002) findings to a college population, the investigators examined co-rumination in heterosexual romantic relationships. Both studies found that there was a significant interaction between relationship and gender (Assad, 2004; Assad & Haley, 2003). While women co-ruminated significantly more than men overall, women were found to co-ruminate significantly more with friends than with partners, whereas men were found to co-ruminate significantly more with partners than with friends. However, one limitation of the previous research on co-rumination is that the data was gathered using questionnaires. While individuals reported these co-rumination patterns, this does not mean that they would actually display these patterns while communicating. Moreover, all of this data was collected from only half of the dyad. Past research did not investigate whether both partners had similar perceptions about the amount of co-rumination in their relationships, nor did past research investigate the importance of the listener in regards to corumination. While there has not been any research on listening behaviors involved with corumination, numerous studies have investigated the importance of the listener in conversations. For example, the listener is an essential component to the conversation because s/he shows the speaker that s/he is being understood (Leaper, Carson, Baker, Holliday, & Myers, 1995). Co-Rumination Communication Patterns 5 Furthermore, Reis and Shaver (1988) argued that the listener is actually more essential to the conversation than the speaker because, by interpreting how the listener responds to disclosed topics, the speaker is able to interpret whether or not s/he is being understood. Additionally, Guerro (1997) examined the nonverbal behaviors of facial pleasantness, smiling, and vocal pleasantness in friendships and romantic partners. Interestingly, Guerro (1997) found a pattern similar to the interaction between gender and relationship found in co-rumination behavior patterns (Assad, 2004; Assad & Haley, 2003). Basically, men were significantly better nonverbal listeners toward their partners than toward their friends, whereas women were significantly better nonverbal listeners toward their friends than toward their partners (Guerro, 1997). The previous research regarding rumination, co-rumination, and listening behaviors lead to three main purposes of the current study. First, a replication of Assad’s (2004) and Assad and Haley’s (2003) study were conducted on co-rumination patterns. Second, listening behavior patterns were examined through videotaped conversations in order to both replicate Guerro’s (1997) findings and to extend them to verbal behaviors. Finally, the co-rumination scores of individuals in the same dyad were correlated to determine if both participants viewed their relationship similarly. The current study examined 8 hypotheses: 1. Rumination scores will significantly correlate with depression scores. 2. Overall, women will co-ruminate significantly more than men. 3. Women will co-ruminate significantly more with friends than with partners. 4. Men will co-ruminate significantly more with partners than with friends. 5. Overall, women will be significantly better listeners than men. 6. Women will be significantly better listeners toward friends than toward Co-Rumination Communication Patterns 6 partners. 7. Men will be significantly better listeners toward partners than toward friends. 8. The co-rumination scores from each member of a dyad will significantly correlate. Method Participants Ninety-six participants (48 women, 48 men) participated in this experiment for extra credit (if they were enrolled in General or Developmental Psychology) and a chance to win 25 dollars. Most of the participants were students at a small, liberal arts college in Minnesota, with 2 participants being high school seniors. Participants were tested in dyads, with 16 dyads consisting of opposite-sex romantic partners, 16 dyads consisting of same-sex female best friends, and 16 dyads consisting of same-sex male best friends. Measures Conversations Questionnaire. The Conversations Questionnaire (Appendix A) collected demographic information and assessed participant’s feelings about both the conversations they participated in that day and about their feelings regarding conversations with their partners in general. Questions included items about how often participants discussed problems/positives in their relationships, how these conversations made them feel, their perceptions of both their and their partner’s listening skills, what they would change about communication within their relationship, and how similar the conversations were to their typical conversations. Coding of conversations. The conversations were rated on several different measures. The listener for each conversation was rated on both verbal and nonverbal measures using 7point Likert scales (Appendix B). The measures rated included boredom/attention (nonverbal), Co-Rumination Communication Patterns 7 rejection/affirmation (nonverbal), boredom/attention (verbal), and negative/positive feedback (verbal). The boredom/attention (nonverbal) category coded for such behaviors as eye contact, leaning forward, and crossing of the arms. The rejection/affirmation (nonverbal) category consisted of measuring behaviors such as smiling or frowning. The boredom/attention (verbal) category included responses such as asking questions, elaboration, and not responding. Finally, the negative/positive feedback (verbal) category measured the listener’s tone, such as by measuring sarcasm or positive affirmation. These variables were tested for interrater reliability by comparing the ratings of 56 coded responses from 2 different raters. The rater’s scores were correlated, and the average of these correlations was computed to obtain interrater reliability, R = .229, p = .095. While the overall correlation was not significant, it was close to being significant. Additionally, since only 56 out of 384 possible variables were included in this analysis, the interrater reliability may improve after including correlations of more data points. Procedure Before starting the experiment, one participant was asked to be “Subject A” and the other was asked to be “Subject B.” The labels were randomly assigned in the romantic partner dyads so that half of the men and half of the women would be “Subject A.” Participants were then seated at a table across from each other. Next, participants signed a consent form and were given instructions by the experimenter. Participants were told they would participate in 3 conversations, all of which would be videotaped. For the first conversation, participants told each other about their days. Next the participants took turns discussing a problem that had happened sometime between the beginning of the school year and the experiment. Finally, participants took turns discussing something positive that had happened between the beginning of the school year and the time of the experiment. For each conversation, Co-Rumination Communication Patterns 8 the experimenter left the room, and then knocked on the door to warn the participants that she was going to give them the next set of instructions. The first conversation lasted for 5 minutes and was unstructured. For the second conversation, the participants were told they should first discuss one of “Subject A’s” problems and would be told by the experimenter when to discuss one of “Subject B’s” problems. Participants discussed “Subject A’s” problem and “Subject B’s” problem for 5 minutes each. A similar procedure was used to discuss participant’s positive experiences, the only difference being that the order was reversed so that “Subject B” had the opportunity to discuss a positive before “Subject A” did. Although participants were focused on only one individual’s problem or positive at a time, both individuals were allowed to talk during all parts of the discussion. After the last conversation, the video-recorder was turned off, and “Subject B” was asked to move to the opposite end of the room. Participants then completed a questionnaire consisting of the Conversations Questionnaire, the Co-Rumination Questionnaire (Rose, 2002), the Friendship Quality Questionnaire (Parker & Asher, 1993), the Depression Scale (Treynor, et al., 2003), and the CES-D Scale (Radloff, 1977). When both participants were finished completing the questionnaires, they were thanked and debriefed. Results Co-rumination and depression Both measures of depression, the CES-D Scale (Radloff, 1977) and Treynor et al.’s (2003) Depression Scale, were correlated with both measures of co-rumination. The Depression Scale was significantly correlated with Rose’s (2002) Co-Rumination Questionnaire, R = .25, p = .019, and was marginally significantly correlated with the Conversations Questionnaire, R = .20, p = .054. Additionally, the CES-D Scale was significantly correlated with the Depression Scale, Co-Rumination Communication Patterns 9 R = .77, p < .001. Furthermore, the Conversations Questionnaire and the Co-Rumination Questionnaire were almost significantly correlated, R = .21, p = .051. While the CES-D Scale did not significantly correlate with either measure of co-rumination overall, it did significantly correlate with the co-rumination scores of romantic partners on the Conversations Questionnaire, R = .62, p < .001. Co-rumination replication A univariate Analysis of Variance (ANOVA) was used to assess the co-rumination variables. In regards to rumination, there was a marginally significant finding that women ruminated more than men, F(1, 91) = 3.60, p = .06. Moreover, in regards to individual variables, the only one that revealed a significant gender difference was that men speculated more on consequences of the problem than did women, F(1, 90) = 4.47, p = .037. While there was not a significant gender difference when all of the variables were combined into an overall corumination variable, there were significant interactions between gender and relationship for many of the individual variables (see Table 1) and an overall interaction in which men coruminated significantly more with partners than with friends and women co-ruminated significantly more with friends than with partners, F(1, 85) = 11.17, p = .001 (see Figure 1). Variable F – value p – value Interaction (M = men, W = women, R = romantic relationship, F = friendship) Frequency discussing problems 9.27 .003 M > R, W > F Frequency encourage other to discuss problems 7.88 .006 M > R, W > F Frequency other encourages discussion of problems 13.46 < .001 M > R, W > F Repeated discussion of problems 6.45 .013 M > R, W > F Speculation on causes of problems 8.08 .006 M > R, W > F Speculation on consequences of problems 4.12 .045 No difference M, W > F Table 1. Significant interactions between gender and relationship on Rose’s (2002) Co-Rumination Questionnaire measures. Co-Rumination Communication Patterns 10 problems Speculation on parts of the problem that are not understood 5.80 .018 M > R, W > F Overall Co-Rumination Scores on the Co-Rumination Questionnaire (Rose, 2002)