Interpersona | An International Journal on Personal Relationships interpersona.psychopen.eu | 1981-6472 Articles Conflict Strategies and Intimacy: Variations by Romantic Relationship Development and Gender

The purpose of the present study was to examine how relationship duration was related to conflict strategies and levels of intimacy in romantic relationships and how that might vary by gender. Participants completed self-report measures to assess perceived levels of reciprocal intimacy and reported uses of positive and negative conflict strategies. Results found an inverted U-shaped pattern for negative conflict strategies and a linear increase in levels of intimacy with duration. No differences were found for positive conflict strategies with duration. Gender differences were found for levels of intimacy, with women reporting higher levels of relationship intimacy as compared to men regardless of relationship duration; but, no gender differences were found for either positive or negative conflict strategies. Future research in this area should replicate these findings to further support the importance of romantic relationship development in examining relationship qualities such as conflict strategies and intimacy.

Interpersona | An International Journal on Personal Relationships interpersona.psychopen.eu | 1981-6472 negative conflict strategies includes behaviors that disrupt the relationship (i.e., blaming and criticizing).For intimacy, the definition proposed by Schwebel, Moss, and Fine (1999) was chosen because of its adherence to the suggestion that intimacy be measured as a transactional variable.Although the concept of love may be unidirectional (e.g., I love him, but he doesn't love me), intimacy requires mutual investment including shared affective, cognitive and physical experiences (Santore, 2008;Schwebel et al., 1999).
Understanding conflict strategies and intimacy in romantic relationships requires some knowledge of the developmental context of romantic relationships.For example, one of the first questions we ask during a conversation with someone about their romantic relationship is "how long have you been dating?"However, few studies examine the developmental context of romantic relationships.Fitness and Planalp (2005) comment that "emotion scholars have tended, inexplicably, to overlook the role of relational context in their theorizing and research" (p.147).
Similarly, a review of articles published in the 1990's found that relationship status was not assessed or was collapsed in most samples suggesting that behaviors and affective states do not change over the development of a relationship (Surra, Boettcher-Burke, Cottle, West, & Gray, 2007).As concluded by the authors of the review, we believe that relationship development does play a role in the behaviors and emotions displayed in a romantic relationship.
An important contextual characteristic of romantic relationship development is the duration of the relationship, which may have important implications for the ways in which conflict and intimacy are experienced (Dainton & Aylor, 2002;Murray & Holmes, 1996;Shulman & Seiffge-Krenke, 2001;Schwebel et al., 1999).Solomon andKnobloch (2001, 2004) have proposed the relational turbulence model which suggests that relationships experience a disruption in previously well-working or smooth patterns of interaction; this disruption is labeled relational turbulence.Knobloch (2007) posits that when a romantic relationship begins the focus is on getting to know one another, but as the relationship continues and becomes more intimate, partners must make predictions about the future of the relationship.The uncertainty surrounding the future of a relationship, or relational uncertainty, has been linked to more negative appraisals of one another, experiences of jealousy, and difficulty communicating with one another.Those couples who are unable to resolve the uncertainty are likely to dissolve the relationship.Those couples who are able to communicate and resolve their feelings of uncertainty are expected to continue the relationship with heightened levels of intimacy and a stronger bond (Knobloch, 2007).Taken together, this model suggests that conflict, over the relational uncertainty, is likely to increase and then fall again (inverted U-shape) whereas levels of intimacy continue to increase over time (linear pattern).
Empirical data for the connection between duration and fluctuations in conflict and intimacy has supported the relational turbulence model.First, Solomon, and Knobloch (2004) found empirical support that levels of intimacy increased over time (linear increase).In addition, Aune, Aune, and Buller (1994) found that romantic partners in relationships of moderate duration reported expressing the most negative emotions.Those in the longest relationships reported less negative emotion than those of moderate duration, but not as low as those from the shortest duration.These results support an inverted U-shape for negative emotions.The general pattern formulated over these studies indicates that intimacy and negative emotions are very low in the early development of romantic relationships.As the relationship progresses both intimacy and negative emotions increase as the couple attempts to label the relationship and determine the future of the relationship.Once the relationship is well-established, negative emotions are expected to decrease whereas levels of intimacy are expected to continue to increase.
Although negative emotions can lead to conflict, the internal experience of negative emotions is not the same as the outward expression of conflict with one's partner.The relational turbulence model suggests that not only will negative emotions be experienced, but conflict over the future of the relationship will also be expressed.Although a large body of research has examined conflict strategies in romantic relationships, few studies have examined how conflict strategies may vary with romantic relationship development.One study did examine both negative and positive conflict strategies during an observational task (Follette & Alexander, 1992).The results indicated that women from relationships of longer duration used more positive conflict strategies than women in relationships of shorter duration; no differences were found for men's positive conflict strategies.In addition, no differences were found for negative conflict strategies and duration for either women or men.However, this study measured duration as a continuous variable and, thus, was unable to examine increases and decreases in nonlinear patterns.
Our study will address this limitation in order to examine fluctuations in positive and negative conflict strategies, as well as levels of intimacy.
The primary purpose of this study was to expand the current knowledge on conflict strategies and levels intimacy in emerging adult romantic relationships within the context of romantic relationship development.Using a crosssectional design, this study examined relationship duration, or the length of time the partners had been together, as an objective measure of romantic relationship development.In order to better capture the nonlinear patterns of fluctuations in conflict strategies, we categorized relationship duration into three groups based on frequencies.
Previous research has used similar procedures for transforming relationship duration into a categorical variable (Aune et al., 1994;Aune & Comstock, 1997;Duemmler & Kobak, 2001;Marcus & Swett, 2002).Our assessment of intimacy in this study was as a reciprocal variable where participants reported on both their own perceptions of the relationship and what they believed to be their partners' perceptions of the relationship.Other researchers have reported on the importance of assessing intimacy in this way (Santore, 2008;Schwebel et al., 1999).Because both theory and empirical data support a linear increase in levels of intimacy (Solomon & Knobloch, 2004), we expected to find the same pattern.Unique to our study is the focus on conflict strategies rather than frequency of conflict or negative emotions.This distinction is important because research suggests that relationship satisfaction is less affected by the frequency of conflict than by the ways in which conflicts are resolved (Berger, McMakin, & Furman, 2005).Hence, our first research question asked:

RQ1: How do positive and negative conflict strategies vary as a function of romantic relationship duration?
Finally, we not only examined the relational context of duration, but also included the individual variable of gender, which has been found to be important in understanding differences in conflict strategies and levels of intimacy.For example, previous research examining conflict strategies with duration found significant differences in positive conflict strategies for women but not men (Follette & Alexander, 1992).In addition, Toomey and Nelson (2001) found emerging adult women report higher levels of intimacy in romantic relationships as compared to men.Hence, our second research question asked:

Sample and Recruitment
Students from a large Midwestern university (N = 297) participated in the study.The sample included 246 women and 51 men with a mean age of 19.7 years (SD = 2.19).Due to the topic of our study, participants were required to be in a current romantic relationship.The sample included mostly Caucasian participants (84.2%) with only 15.8% of participants reporting non-white ethnicities including 9.4% African-American participants, 2.7% Hispanic/Latino participants, 1.3% Asian-American participants, and 2.0% participants of other ethnicities.Participants were raised in a variety of family structures including intact families (73.4%), divorced families (19.9%), singleparent families (3.0%), and other types of family arrangements (3.7%).Participants reported that their mothers' education level varied from 2.7% who did not graduate high school, 29.3% who had a high school diploma, 24.6% who had some college education, 32.0% who had a four year degree, and 11.4% who had a graduate level degree.
For fathers' education level, participants reported 3.7% did not graduate high school, 29.0% had a high school diploma, 15.8% had some college education, 34.7% had a four year degree, and 14.8% had a graduate level degree.

Measures
Romantic Relationship Development.-We used duration of the relationship as a proxy for romantic relationship development.Participants reported the number of months they had been in their current romantic relationship.
The mean duration was 18.28 months (SD = 15.64) with a range of one month to seven years.We could find no theoretical or empirical data to provide guidance on how to use duration to create categories or groups (e.g., the literature does not inform us about the typical time a couple is in a relationship prior to experiencing relational uncertainty).Following many other studies (Aune et al., 1994;Aune & Comstock, 1997;Duemmler & Kobak, 2001;Marcus & Swett, 2002), a three-way split based on frequencies was used to create three groups (short, moderate, and long).Participants who reported a relationship duration of seven months or less (N = 94) were included in the short duration group.Those who reported durations of eight to 23 months (N = 96) were included in the moderate duration group, and those who reported durations of 24 months or longer (N = 97) were included in the long duration group.
Conflict Strategies.-Conflict strategies within romantic relationships were assessed using the 109-item Managing Affect and Differences Scale (MADS; Arellano & Markman, 1995).For each item, participants indicated how much they agreed with each statement using a Likert scale from 1 = strongly disagree to 5 = strongly agree.The measure assesses both positive (e.g., emotional expressivity, validation, feedback, and leveling) and negative (e.g., negative conflict escalation, negativity, and conflict withdrawal) conflict strategies.The MADS has good internal consistency and negative conflict scales have been validated against actual behavior observations (Arellano & Markman, 1995).In the present study, alpha coefficients were .95for the positive subscale and .91 for the negative subscale.
Intimacy.-Intimacy in romantic relationships was assessed using a revised version of the Miller Social Intimacy Scale (MSIS; Miller & Lefcourt, 1982).This scale includes seventeen original items that assess expressed intimacy and an additional seventeen items that were added to reflect the reciprocity of intimacy, or received intimacy.
Items are answered on a 10-point Likert scale.Twelve of the items assess frequency (i.e., "How often do you confide very personal information to him/her?" and "How often does he/she confide very personal information to you?") rated from 1 = very rarely to 10 = almost always; and 22 items to assess intensity (i.e., "How close do you feel to him/her most of the time?" and "How close does he/she feel to you most of the time?") rated from 1 = not much to 10 = a great deal.All items are summed to create a total intimacy score that can range from 34 to 340 with higher scores indicating a higher level of intimacy.In a previous study, the internal consistency as estimated by the Cronbach's alpha coefficient was .91,and test-retest reliabilities were measured as r = .84after a two month interval.Convergent, discriminant, and construct validity were also demonstrated.In the present study, the alpha coefficient was .89for the total intimacy score.

Procedures
Data collection included group testing sessions where participants gathered in a large lecture hall.During this time, informed consent was obtained and the paper-and-pencil measures were completed.Measures were counterbalanced.Following testing, participants were given short written debriefing statements.All of the participants were compensated for their time with extra credit points for an undergraduate psychology course.

Results
A 2 x 3 ANOVA was used to assess group differences for reports of positive conflict strategies based on gender and relationship duration (short, moderate, long).A 2 x 3 ANOVA revealed no main effect for gender, F(1, 287) = 1.89, p = .17,partial η 2 = .01,no main effect for relationship duration, F(2,287) = 2.75, p = .07,partial η 2 = .02,and no interaction between the two variables, F(2, 287) = 2.98, p = .05,partial η 2 = .02(see Table 1).A second 2 x 3 ANOVA was used to assess the same group differences (gender and relationship duration) for reports of negative conflict management.Although there was no main effect for gender, F(1,292) = .92,p = .34,partial η 2 = .00,and no interaction effect, F(2,292) = 2.86, p = .06,partial η 2 = .02,there was a main effect for relationship duration, F(2,292) = 4.34, p < .01,partial η 2 = .30(see Table 1).Post hoc test revealed that participants who were in the moderate duration group reported more negative conflict as compared to those in the short duration group (p = .01)and long duration group (p = .02).No mean differences were found between the short and long duration groups.Therefore, an inverted U-shape was found for negative conflict strategies (see Figure 1).
Another 2 x 3 ANOVA was used to assess the same group differences (gender and relationship duration) for reports of intimacy levels.There was a main effect for gender, F(1,291) = 9.20, p < .01,partial η 2 = .03,with women reporting higher levels of intimacy (M = 294.96)than men (M = 283.72).A main effect was also found for relationship duration, F(2,291) = 4.40, p = .01,partial η 2 = .03(see Table 1).Participants who were in the long duration group reported higher levels of intimacy than those in the short duration group (p < .01).No significant differences were found between the short and moderate duration groups or between the moderate and long duration groups (see Figure 2).The interaction effect was not significant, F(2,291) = 2.15, p = .12,partial η 2 = .02.

Discussion
The purpose of this study was to add to the existing literature on conflict strategies and intimacy in emerging adult romantic relationships.Specifically, we examined the context of relationship duration, as well as gender, for differences in conflict strategies and intimacy.Our results indicated no differences in positive strategies across relationship duration.Negative conflict strategies were found to follow an inverted U-shape which is consistent with the relational turbulence model (Solomon & Knobloch, 2004) and research on how negative emotions fluctuate (Follette & Alexander, 1992).Finally, and consistent with previous research, levels of intimacy were found to have a linear increase with duration and be reported higher among women than men.
The finding that positive conflict strategies did not significantly vary by romantic relationship duration was somewhat surprising.Conflict, negative emotions, and intimacy have all been found to fluctuate in previous research and our own data found that negative conflict strategies vary by relationship duration.The lack of variation may be that positive strategies are initially used when one is still attempting to impress the other with high levels of positive self-presentation.As the relationship develops, positive conflict strategies may become habitual or routine.For example, Dainton and Aylor (2002) found that relationship duration was positively related to the routine use of positive conflict strategies.These researchers define routine use as the unintentional use of these strategies to maintain the relationship, as opposed to the strategic use which includes a more conscious effort to maintain the relationship.Hence, positive strategies may not fluctuate even though the purpose of those behaviors may change.
Our finding that negative conflict strategies increase (higher among moderate versus short duration) and then decrease (lower among long versus moderate duration) is consistent with other research that found higher reports of negative emotions at moderate durations of romantic relationships than shorter or longer durations (Aune et al., 1994).Our finding are unique because rather than focusing on internal emotional states that may or may not be expressed, we focus on conflict behaviors or strategies.Our findings also provide additional support for the relational turbulence model which suggests that at one or more points during the development of a romantic relationship partners will experience relational uncertainty (Knobloch, 2007;Solomon & Knobloch, 2001, 2004).The uncertainty surrounding the future of a relationship is likely to result in more negative appraisals of one another, experiences of jealousy, and difficulty communicating with one another-all of which are likely to contribute to an increase in negative conflict strategies.Establishing a mutual understanding about the future of the relationship will decrease uncertainty and, according to our data, decrease the amount of negative conflict strategies used in the relationship.Future research is needed that examines more closely if there is a connection between relational  uncertainty and the specific issues that spark conflict (e.g., are issues of jealousy more likely to spark conflicts in relationships of short, moderate, or long duration?).
Intimacy levels were expected to increase across the three relationship durations.Significant increases were found between the short and long duration groups, but not between the short and moderate durations or between the moderate and long durations.This finding suggests a linear, yet gradual, progression of intimacy across romantic relationship development.Our results are consistent with Solomon and Knobloch's (2004) findings of a linear increase in intimacy over time.
Although some studies have found gender differences in conflict strategies (Feldman & Gowen, 1998;Hojjat, 2000;Schwarzwald & Koslowsky, 1999), gender differences were not found for either positive or negative conflict strategies in the current study.This finding supports theory and research suggesting that the gender gap may be closing in some areas of relationship development (Shulman, Levy-Shiff, Kedem, & Alon, 1997).It may also be that men and women differ in the types of strategies used rather than their use of positive or negative strategies in general.Using negative strategies as an example, research has found that men are more likely to use coercion and women were more likely to express negative emotions (Keashly, 1994).In addition, women were more likely to sulk and criticize while men were more likely to show anger and avoid discussions (Peplau, 1983).Future research should further examine gender differences in specific conflict strategies and how those may vary with relationship development.
As expected, gender differences did emerge for levels of intimacy with women reporting higher levels of intimacy than men.This finding is consistent with studies that found higher reports of intimacy among women as compared to men (Montgomery, 2005;Toomey & Nelson, 2001), yet our finding is in contrast to research suggesting a narrowing in the gender gap with regard to intimacy (Shulman et al., 1997).Because intimacy can be measured in many different ways (i.e., feelings of closeness, disclosure of personal information), future research should assess levels of intimacy with multiple measures to determine if similar gender differences exist with other measures.Some limitations should be noted.First, this study was comprised entirely of college students and cannot be generalized to non-college populations.It is possible that certain aspects of college life affect the development of a relationship (i.e., more flexible schedules, long holiday breaks, etc.).Second, information was not obtained by

Figure 1 .
Figure 1.Differences in negative conflict strategies by relationship duration.

Figure 2 .
Figure 2. Differences in levels of intimacy by relationship duration.
both partners in the relationship nor were multiple measures used to assess each of the constructs.Third, romantic relationship development was assessed with arbitrary durations.Although duration is more objective than subjective judgments of seriousness, multiple measures of romantic relationship development are needed to determine a better understanding of the normative patterns of romantic relationship development.For example, when does the need to define the relationship and decrease relational uncertainty begin?How long does this period of relational turbulence typically last in the development of a romantic relationship?What might be some important milestones (e.g., saying I love you) that signal the beginning of a period of turbulence?In order to adequately address these questions, not only are better measures of romantic relationship development needed, but so are studies that use longitudinal designs.Our cross-sectional design does not allow us to investigate change over time.Longitudinal studies that follow the development of romantic relationships over time, including changes in conflict and intimacy, are needed to more rigorously examine are research questions.Obviously, variations in romantic relationship development exist, but normative patterns of development over time could be discovered to help understand vulnerable periods of romantic relationships in emerging adulthood.The present study adds to current literature on the importance of examining romantic relationship development during emerging adulthood.Our findings suggest that variation in results between studies on romantic relationships may be due to different phases of relationships being assessed.Relationships in the early phases often involve a larger amount of affiliative behaviors (i.e., having fun, sharing interests, and spending time together), which mostly involve positive affect and usually do not involve strong emotions or highly sensitive communication topics(Berger et al., 2005).Studies where a majority of the relationships assessed are new or less committed are likely Interpersona 2012, Vol.6(2), 200-210 doi:10.5964/ijpr.v6i2.101

Table 1
Means and Standard Deviations for Conflict Strategies and Intimacy by Gender and Relationship Duration