On the Benefits of Valuing Being Friends for Non-Marital Romantic Partners

Romantic relationships are, at their core, friendships. As such, it may be the case that valuing that aspect of the relationship fortifies the romantic relationship against negative outcomes and serves as a buffer against dissolution. We explored the role of valuing friendship within romantic relationships in two two-wave studies examining whether investing in the friendship aspect of the relationship (study 1; N = 190) and placing importance on affiliative need fulfillment (study 2; N = 184) were associated with positive concurrent outcomes and positive outcomes over time. Results revealed that valuing the friendship aspect of a romance is a strong positive predictor of concurrent romantic relationship qualities (i.e., love, sexual gratification and romantic commitment), is associated with increases in these qualities over time, and is negatively associated with romantic dissolution. Furthermore, evidence suggests that these benefits come from valuing friendship specifically, rather than any other aspect of the relationship (e.g., the sexual aspect). The Benefits of Valuing Being Friends 3 On the Benefits of Valuing Being Friends for Non-Marital Romantic Partners Committed romantic relationships are a central aspect of human life. Such relationships have the potential to fulfill individuals’ fundamental needs for belonging and social affiliation, but maintaining them over time can be challenging. When relationships dissolve, the former partners experience myriad negative outcomes, including negative emotions, insecurity, and decreased physical health (see Fine & Harvey, 2006). Understanding the causes of decreasing relationship quality and dissolution has the potential to help avoid or prepare for these negative outcomes. Much is known about the proximal predictors of declining relationship quality, including what cognitions and behaviors are indicative of a relationship that is likely to dissolve (Le, Smoak, Agnew, Korn, & Mutso, 2010). In the current studies we sought to add to what is known about the predictors of romantic relationship outcomes (e.g., commitment, dissolution), arguing that something as fundamental and potentially distal to the outcomes as which qualities of their relationships the partners value most highly can have a robust impact on relationship quality over time and the likelihood of relationship dissolution. Romantic relationships, despite being categorized under the one label, can vary qualitatively from couple to couple. Given their ability to fulfill a wide array of individuals’ needs, from the need to affiliate with others to sexual needs and beyond, romantic relationships offer truly endless variations. Indeed, “different people approach and plan their romantic lives in distinct ways that is, in ways that fit with their overarching needs and goals” (p. 241, Sanderson, Keiter, Miles, & Yopyk, 2007). How much value the partners place on each of the needs, we argue, influences the relationship outcomes they attain. We hold that the value placed on one set of needs in particular, friendship with a romantic partner, is responsible for notable positive relationship outcomes. The Benefits of Valuing Being Friends 4 The Importance of Friendship in Romance There is some divergence among theorists and theories about the exact nature of friendship, but most agree that it is characterized by the importance of affection, intimacy, reliable alliance, and instrumental and emotional support (Berndt, 2002). Friendships are considered an integral part of human existence, serving numerous functions, including forming the basis of romantic relationships (Fehr, 1996). Evidence that friendship is central to romantic relationships comes from many sources, but perhaps the most informative is the evidence derived from prototype analyses of romantic constructs, as these provide insight into what lay people view as central aspects. In several analyses of this type, friendship emerged as a central, defining characteristic of love, broadly defined (Fehr, 1988, 1994; Fehr & Russell, 1991), of romantic love, specifically (Regan, Kocan, & Whitlock, 1998), and of romantic relationships (Aron & Westbay, 1996). Not only is friendship central to the concepts of love and romantic relationships, but valuing and experiencing friendship in romances is associated with receiving positive outcomes. Those individuals who conceptualize friendship qualities as important to love experience greater friendship with their romantic partners (Aron & Westbay, 1996), and importantly, greater overall relationship quality (Fletcher & Kininmonth, 1992; Fletcher, Rosanowski, & Fitness, 1994). Additionally, the actual experience of greater friendship within a romantic relationship is associated with experiencing greater passion in the romance (Grote & Frieze, 1994), and satisfaction with the romance and life generally (Fehr, 1996; Kim & Hatfield, 2004). Additional research is needed to determine whether the benefits of valuing friendship within a romance yields these benefits over time, and whether these benefits translate into a lower likelihood of relationship dissolution. The Benefits of Valuing Being Friends 5 That valuing the friendship in a romance is associated with a greater experience of passion in that romance raises the question of whether friendship is uniquely suited to providing these positive outcomes for the relationship, or if simply valuing any aspect of the relationship (e.g., the passionate aspect), would yield the same benefits. There is some evidence suggesting the former, that friendship is more highly responsible for the positive outcomes than is passion. In prototype analyses, the friendship aspect of the relationship is rated as more central to the definition of a romantic relationship and romantic love than are the passionate and sexual aspects (Aron & Westbay, 1996; Regan et al., 1998). Interestingly, those individuals who place greater importance on passion do not report a greater experience of passion than those who place a lesser importance on it (Aron & Westbay, 1996), and a belief that passion is important is unassociated with romantic relationship quality (Fletcher & Kininmonth, 1992). Taken together, evidence suggests that valuing the friendship aspect of a romantic relationship is associated with positive outcomes for the relationship that seem to not derive from valuing other aspects of the relationship. Nevertheless, additional research directly comparing the benefits of valuing friendship and valuing other aspects of the relationship is needed to ascertain whether valuing friendship in a romance provides relational benefits uniquely from valuing other aspects of the relationship. The Current Studies Our goal in the current research was to develop a greater understanding of the importance of valuing friendship to relationship functioning, both concurrently and over time, and both independently and after controlling for the value placed on other aspects of the relationship. To that end, we conducted two studies, employing different measures of valuing the friendship in each. The Benefits of Valuing Being Friends 6 In Study 1, we assessed the behavior people enact when they value their relationships and want them to continue. Specifically, we measured the amount of valued linkages individuals have made or plan to make into the friendship aspect of their relationship (i.e., friendship investment). These valued linkages include investments made in the past, as well as the investments individuals plan to make in the future. In both cases, the investments are the resources that might be lost if the relationship were to end (Agnew, Arriaga, & Wilson, 2008; Goodfriend & Agnew, 2008; Rusbult, 1980; Rusbult, Martz, & Agnew, 1998), including both tangible (e.g., jointly owned furniture, children, shared debt), and intangible resources (e.g., selfdisclosures, time, effort; Goodfriend & Agnew, 2008). By measuring the behavior of making investments, we can infer how much individuals value their friendship, which we hypothesized would be associated with both concurrent relationship quality, as well as relationship quality over time: Friendship investment will be significantly and positively associated with concurrent romantic commitment (Hypothesis 1a), love (Hypothesis 2a) and sexual need fulfillment (Hypothesis 3a). Furthermore, friendship investment at Time 1 will be associated with positive change in romantic commitment by Time 2 (Hypothesis 1b), positive change in love by Time 2 (Hypothesis 2b), and positive change in sexual need fulfillment by Time 2 (Hypothesis 3b). Finally, friendship investment at Time 1 will be negatively associated with leave behavior (i.e., steps taken to leave one’s romantic relationship reported at Time 2; Hypothesis 4). A benefit of examining the behavior of investment is that it allows us to assess individuals’ values in a way that does not require them to know precisely how important they think the friendship is. Nevertheless, it is a proxy for the construct of interest. In Study 2, we wanted to conceptually replicate the findings in Study 1 with a more direct measure of value, and The Benefits of Valuing Being Friends 7 extend the findings by comparing the relational outcomes associated with valuing the friendship aspect of the relationship to those associated with valuing other aspects. Taking a need fulfillment approach (e.g., Drigotas & Rusbult, 1992; Fehr, 1996) is most appropriate here, as we can aggregate the needs romantic relationships fulfill into categories and compare those that are defining characteristics of friendships (i.e., affiliative needs such as companionship and support; Fehr, 1996) from those that are not (i.e., sexual needs, personal needs). We expected that valuing the friendship aspect of the relationship would predict the relational outcomes above and beyond the effect of valuing other aspects, and that placing greater importance on the friendship aspect of the relationship relative to other aspects would yield positive relational outcomes and promote relationship persistence: The importance placed on affiliative need fulfillment will be significantly and positively associated with concurrent commitment (Hypothesis 5a) love (Hypothesis 6a) and sexual need fulfillment (Hypothesis 7a) above and beyond the importance of personal and sexual needs, but the importance placed on personal needs and sexual needs will not be significantly associated with concurrent commitment, love or sexual fulfillment above and beyond the importance of affiliative needs (Hypothesis 8a). Additionally, the importance placed on affiliative needs will be associated with positive changes in romantic commitment (Hypothesis 5b), love (Hypothesis 6b), and sexual need fulfillment (Hypothesis 7b) by Time 2, but the importance of personal needs and sexual needs will not be significantly associated with these changes above and beyond the importance of affiliative needs (Hypothesis 8b). Finally, those who endorsed that fulfillment of affiliative needs was most important (compared to personal and sexual needs) at Time 1 will be significantly less likely to have engaged in leave behavior by Time 2 than will those who endorsed either personal or sexual need fulfillment as most important (Hypothesis 9). The Benefits of Valuing Being Friends 8 Together, these studies provide a multifaceted look at the idea of valuing the friendship aspect of a relationship, encompassing both the behaviors in which individuals engage, and the cognitions they have when they value the relationship. Study 1 Method Design and Participants. This study employed a two-wave design. Approximately four months after participation at Time 1 participants were contacted and asked to complete a followup questionnaire. Undergraduate students at Purdue University who met the eligibility requirement of being in a heterosexual romantic relationship at Time 1 and who participated at Time 2 were included in the analyses (N = 190; 76 males and 114 females). The average duration of their relationship at Time 1 was 17.8 months (SD = 16.6, Median = 13). Most participants indicated that they were involved in an exclusive dating relationship (92.0 % with 8.0% casually dating). Participants’ ages ranged from 18 to 27 years (M = 19.34, SD = 1.22), and the majority indicated that they were White (87.0%, with 4.2% Asian, 3.7% Black, 3.7% Hispanic, and 1.4% indicating other). At Time 2, 51 (27.0%) of the participants indicated they were no longer dating their Time 1 romantic partner while 138 indicated they were still romantically involved with their Time 1 partner. Procedure. Participants signed up for a particular time to complete the Time 1 portion of the study through the Purdue University subject pool website. All participants completed the measures described below in partial fulfillment of an introductory psychology course requirement. They completed the measures in large computer labs across campus, after which they were debriefed and thanked for their time. The Benefits of Valuing Being Friends 9 Approximately four months after participating at Time 1 (M = 4.29, SD = .37), participants were contacted via email individually and invited to return to the questionnaire web site to complete a Time 2 questionnaire. They were reminded of their Time 1 partner’s first name prior to completing the measures described below. Participants were allowed to complete Time 2 measures at whatever time they chose from any location with Internet access. Time 1 Measures. At Time 1, all participants completed six items assessing valued linkages (i.e., past and planned investments) to the friendship within their current relationship. Recent work has shown that the investment construct, which has been described as the resources already linked to the relationship, is more predictive of commitment when considered as the combination of past tangible and intangible resources as well as tangible and intangible resources than the dyad plans to invest in the relationship in the future (Agnew, Arriaga, & Wilson, 2008; Goodfriend & Agnew, 2008). Thus, both past and planned investments in the friendship were assessed. Participants were asked to answer three past investment items with regard to the friendship component of their romantic relationship. These items were adapted from the Investment Model Scale (Rusbult, Martz, & Agnew, 1998) to apply specifically to the friendship component (e.g., an item that previously read “I have put a great deal into our relationship that I would lose if the relationship were to end,” was adapted to read “I have put a great deal into our friendship that I would lose if the relationship were to end”). The complete valued linkage scale includes the items: “I feel very involved in our friendship -like I have put a great deal into it,” and “Compared to other people I know, I have invested a great deal in my friendship with my partner.” The three planned investment items were adapted from the Investment Model Scale (Agnew, Arriaga, & Wilson, 2008; Rusbult et al., 1998), to apply specifically to the friendship component of the romantic relationship as well as to refer to the future: “In the future, I plan to The Benefits of Valuing Being Friends 10 put a great deal into our friendship that I would lose if the relationship were to end,” “In the future, I plan to be very involved in our friendship -to put a great deal into it,” and “In the future, I plan to invest a great deal in my friendship with my partner compared to other people I know.” These items used a nine-point response scale ranging from 1 (“do not agree at all”) to 9 (“agree completely”). Reliability of this friendship investment scale was high (α = .90). Additionally at Time 1, participants completed the seven-item Investment Model commitment scale (Rusbult, et al., 1998) adapted to apply specifically to the romantic relationship (e.g., an item that previously read “I want our relationship to last a very long time” was adapted to read “I want our romantic relationship to last a very long time”). These items employed a nine-point response scale ranging from 1 (“do not agree at all”) to 9 (“agree completely”). Consistent with past findings with the Investment Model Scale, the reliability of the commitment scale was high (α = .92). Finally, at Time 1, participants also completed one item to assess how in love they were with their partner (“How in love are you with your current partner?”) and one item to assess how fulfilled their sexual needs were within their relationship (“My needs for sexual contact (having physical intimacy) are fulfilled by my relationship with my current partner”). Both of these items employed a nine-point response scale ranging from 1 (“not at all” / “not at all fulfilled”) to nine (“very much” / “completely fulfilled”). Time 2 Measures. At Time 2 all participants were asked the following question to assess continued involvement in their romantic relationship: “Are you still romantically involved with this person?” Possible responses were “No, we are not romantically involved (i.e., we broke up)” and “Yes, we are still romantically involved.” As has been done in previous research (VanderDrift, Agnew, & Wilson, 2009), participants who answered “no” completed a subscale The Benefits of Valuing Being Friends 11 from the Assessment of Relationship Changes (Agnew, Arriaga, & Goodfriend, 2006). The relevant subscale included one question regarding deciding to leave the relationship ("In the end, who made the final decision to end your romantic relationship?"), one question regarding initiating dissolution ("Who was the person who first said something or did something that initiated the end to your romantic relationship?") and one question regarding suggesting dissolution ("Who first suggested ending your romantic relationship?"). The response options for these items were "You" or "Your Partner," and were coded such that "You" was 1 and "Your partner" was 0. The mean of these three items was calculated and used as an overall action index, ranging from 0 (no action taken toward dissolution by participant) to 1 (all actions taken toward dissolution by participant). Participants who reported that their relationship had not ended by Time 2 were assigned a 0 on the action index, as having not dissolved indicated no action toward dissolution was taken. Combining the three individual leave behaviors formed a reliable scale (α = .88). Additionally, all participants, regardless of whether their relationship was intact or dissolved, completed a three-item version of the commitment scale used at Time 1, including the items: “I want our romantic relationship to last a very long time,” “I am committed to maintaining my romantic relationship with this person,” and “I feel very attached to our romantic relationship.” These items employ a nine-point response scale ranging from 1 (“do not agree at all”) to 9 (“agree completely”). The reliability of these items was high (α = .99). They also completed one item to measure how in love they were with their Time 1 partner (“How in love are you with this person now?”) and one to measure how fulfilled their sexual needs were by their Time 1 partner (“My needs for sexual contact (having physical intimacy) are fulfilled by The Benefits of Valuing Being Friends 12 my relationship with this person.”) These items employed a nine-point response scale ranging from 1 (“not at all” / “not at all fulfilled”) to 9 (“very much” / “completely fulfilled”). Results Relationship duration at Time 1 was used as a covariate in all tests of hypotheses, as it is significantly (negatively) associated with leave behavior. The pattern of results is identical to what is presented when this covariate is not included (i.e., all coefficients that were significant with it are significant without it and all coefficients that were non-significant with it remain nonsignificant without it). Descriptive analyses of study measures. At Time 1, high levels of friendship investment (M = 7.13; SD = 1.58), romantic relationship commitment (M = 7.42; SD = 1.71), love (M = 7.59; SD = 1.77), and sexual need fulfillment (M = 7.88; SD = 1.63) were reported by participants. At Time 2, moderate levels of romantic relationship commitment (M = 6.56; SD = 3.13), love (M = 6.70; SD = 2.71), and sexual need fulfillment (M = 6.36; SD = 2.98) were reported. Testing Hypotheses 1a-3a. We hypothesized that friendship investment would be significantly and positively associated with concurrent romantic commitment (Hypothesis 1a), love (Hypothesis 2a) and sexual need fulfillment (Hypothesis 3a). Results from multiple regression analyses provide support for each of these hypotheses, indicating that friendship investment is significantly associated with romantic commitment (β = .62, t(187) = 10.61, p < .001), love (β = .61, t(187) = 10.46, p < .001), and sexual need fulfillment (β = .31, t(187) = 4.40, p < .001). Testing Hypothesis 1b-3b. We further hypothesized that friendship investment at Time 1 would be associated with positive change in romantic commitment by Time 2 (Hypothesis 1b), The Benefits of Valuing Being Friends 13 positive change in love by Time 2 (Hypothesis 2b), and positive change in sexual need fulfillment by Time 2 (Hypothesis 3b). To test these hypotheses, we conducted multiple regression analysis in which friendship investment was used to predict each outcome at Time 2 (i.e., romantic commitment, love, sexual need fulfillment) controlling for the effect of the same outcome at Time 1. Results supported our hypotheses, indicating that friendship investment is positively associated with increases in romantic commitment (β = .32, t(186) = 3.76, p < .001), love (β = .19, t(186) = 2.43, p = .016), and sexual need fulfillment (β = .24, t(186) = 3.19, p = .002). Testing Hypothesis 4. Hypothesis 4 holds that friendship investment at Time 1 will be negatively associated with self-reported leave behavior. Results supported this hypothesis, indicating that friendship investment is significantly, negatively associated with individuals reporting that they have engaged in leave behavior (β = -.17, t(187) = -2.32, p = .021). Discussion Taken together, the results of Study 1 indicate that valuing the friendship aspect of a romantic relationship is important to relationship quality both concurrently and over time. Extending these findings, it seems likely that placing greater importance on the friendship component of the relationship relative to other components (e.g., sex) may promote relationship persistence. Before making this extension, however, there are two issues to consider regarding Study 1. The first issue to consider prior to concluding that increased importance paid to the friendship component of a relationship is associated with increased positive outcomes is that the measures used in this study did not directly assess how important friendship is to participants. Instead, the measures assessed the self-reported behavior in which individuals who value the The Benefits of Valuing Being Friends 14 friendship are more likely to engage (i.e., partners who value friendship in a relationship are more likely to invest in the friendship). To complement this self-reported behavioral measure, in Study 2 we collected a measure that directly assesses the importance placed on the friendship aspect of the relationship. The second issue to consider is that the measures used in Study 1 assess the impact of valuing the friendship aspect of a relationship, but fail to assess other potentially important components of the relationship (e.g., the sexual component). Results from Study 1 provide evidence that friendship is important to relationship outcomes, but it may be that, contrary to our hypotheses, placing value on any part of the relationship is sufficient to generate positive relationship outcomes. Accordingly, in Study 2 we collected measures of the importance of friendship as well as of other aspects of the relationship to see if valuing friendship is uniquely associated with positive outcomes. Study 2 Method Design and Participants. This study also employed a two-wave design. Approximately four months after participation at Time 1 participants were contacted and asked to complete a follow-up questionnaire. Undergraduate students at Purdue who met the eligibility requirement of being in a romantic relationship at Time 1 and who participated at Time 2 were included in the analyses (N = 184; 78 males and 106 females). The average duration of their relationship at Time 1 was 15.79 months (SD = 13.2, Median = 12). Most participants indicated that they were involved in an exclusive dating relationship (96.2% with 3.8 % casually dating). Participants’ ages ranged from 18 to 30 years (M = 19.42, SD = 1.40), and the majority indicated that they were White (90.5%, with 2.6% Asian, 2.6% Black, 2.6% Hispanic, and 1.6% indicating other). The Benefits of Valuing Being Friends 15 At Time 2, 54 (29.5%) of the participants indicated they were no longer dating their Time 1 romantic partner whereas 129 indicated they were still involved with their Time 1 partner. Procedure. Participants signed up for a particular time to complete the study through the Purdue University subject pool website. All participants completed the measures described below in partial fulfillment of an introductory psychology course requirement. They completed the measures in large computer labs across campus, after which they were debriefed and thanked for their time. Approximately four months after participating at Time 1 (M = 4.34, SD = .47), participants were contacted via email individually and invited to return to the questionnaire web site to complete a Time 2 questionnaire. They were reminded of their Time 1 partner’s first name prior to completing the measures described below. Participants were allowed to complete Time 2 measures at whatever time they chose from any location with Internet access. Time 1 Measures. The importance of need fulfillment was assessed by asking each participant to indicate how important it is to them that the each of seven specific needs be fulfilled in their relationship with their current partner on a scale from 1 (“not at all important”) to 9 (“extremely important”) (see Drigotas & Rusbult, 1992). The needs examined were: companionship (sharing time and activities), sexual contact (having physical intimacy), security (feeling supported, protected), care-giving (giving support, protection), self-expansion (having new and exciting experiences), self-improvement (experiencing personal growth), and independence (having my own space and making my own decisions). The name of each need and the brief description in parentheses following the need name were both presented to ensure each need was uniformly understood by the participants. Responses to the seven need importance items were subjected to principal component analysis (PCA) to reduce the number of items into The Benefits of Valuing Being Friends 16 meaningful groups. The principal component method was used to extract the components followed by a Promax (oblique) rotation. Three components evidenced Eigenvalues greater than 1 and a Scree plot also suggested the presence of three meaningful components. Accordingly, three components, explaining 69% of the total variance, were retained for rotation. In interpreting the rotated component pattern, an item was held to load on a given component if the loading was greater than .60 on that component and less than .40 for any other. Using these criteria, three needs were found to load on the first component, which was titled “Affiliative Needs” (i.e., companionship, care-giving, and security). Three items were also found to load on the second component, which was titled “Personal Needs” (i.e., self-expansion, selfimprovement, and independence). The final need was found to load on a third component, which was titled “Sexual Needs” (i.e., sexual contact). See Table 1 for complete PCA results. Participants also completed the same one item assessing how in love they were with their partner, one item assessing how fulfilled their sexual needs were within their relationship, and the seven-item commitment subscale of the Investment Model Scale adapted to apply specifically to the romantic relationship as used in Study 1 (commitment α = .91). Time 2 Measures. At Time 2 all participants were asked the following question to assess relationship stability: “Are you still romantically involved with this person?” Possible responses were “No, we are not romantically involved (i.e., we broke up)” and “Yes, we are still romantically involved.” Participants who reported their relationship had dissolved then answered the same subscale from the ARC (Agnew, Arriaga, & Goodfriend, 2006) as used in Study 1 to assess their leave behavior. As in Study 1, combining the three individual leave behaviors formed a reliable scale (α = .87). The Benefits of Valuing Being Friends 17 Additionally, all participants, regardless of whether their relationship was intact or dissolved, completed the same outcome measures as used in Study 1: a three-item version of the commitment scale (α = .91), one item to measure how in love they were with their Time 1 partner, and one item to measure how fulfilled their sexual needs were by their Time 1 partner. Results As in Study 1, all hypotheses were tested including relationship duration at Time 1 as a covariate. The pattern of results is identical to what is presented when this covariate is not included (i.e., all coefficients that were significant with it are significant without it and all coefficients that were non-significant with it remain non-significant without it). Descriptive analyses of study measures. At Time 1, high levels of importance was placed on the fulfillment of affiliative needs (M = 8.08; SD = .99), personal needs (M = 7.83; SD = .98), and sexual needs (M = 7.36; SD = 1.68). Additionally, high levels of romantic commitment (M = 7.49; SD = 1.61), love (M = 7.66; SD = 1.75), and sexual need fulfillment (M = 7.84; SD = 1.78) were also reported by participants at Time 1. At Time 2, moderate levels of romantic relationship commitment (M = 6.58; SD = 3.15), love (M = 6.64; SD = 2.88), and sexual need fulfillment (M = 6.22; SD = 3.16) were reported. Testing Hypotheses 5a-7a. We hypothesized that the importance of affiliative needs will be significantly and positively associated with concurrent romantic commitment (Hypothesis 5a), love (Hypothesis 6a), and sexual need fulfillment (Hypothesis 7a) above and beyond the importance of personal and sexual needs. Results from multiple regression analyses controlling for the importance of personal and sexual needs provide support for each of these hypotheses, indicating that the importance of affiliative needs is significantly associated with commitment (β The Benefits of Valuing Being Friends 18 = .42, t(179) = 5.87, p < .001), love (β = .42, t(179) = 5.85, p < .001), and sexual need fulfillment (β = .16, t(179) = 2.15, p = .033). Testing Hypothesis 8a. Hypothesis 8a holds that the importance of personal needs and sexual needs will not be significantly associated with concurrent romantic commitment, love, or sexual need fulfillment above and beyond the importance of affiliative needs. Results from multiple regression analysis provided partial support for this hypothesis. Controlling for the importance of affiliative and sexual needs, personal needs are not significantly associated with love (β = -.11, t(179) = -1.53, p > .13) or sexual need fulfillment (β = -.12, t(179) = -1.53, p > .13), and whereas there is a significant association between the importance of personal needs and commitment, it is a negative association with commitment (β = -.15, t(179) = -2.07, p = .040). Controlling for the importance of affiliative and personal needs, sexual needs are also not significantly associated with commitment (β = -.09, t(179) = -1.25, p = .21) or love (β = .00, t(179) = 0.02, p = .99), but are significantly and positively associated with sexual need fulfillment (β = .20, t(179) = 2.81, p = .005). Testing Hypothesis 5b-7b. We hypothesized that the importance placed on affiliative needs at Time 1 would be associated with positive change in romantic commitment by Time 2 (Hypothesis 5b), positive change in love by Time 2 (Hypothesis 6b), and positive change in sexual need fulfillment by Time 2 (Hypothesis 7b) above and beyond the effect of the importance placed on personal or sexual need fulfillment. To test these hypotheses, we conducted multiple regression analysis in which the importance of affiliative needs was held to predict the outcome at Time 2 (i.e., romantic commitment, love, sexual need fulfillment) controlling for the effect of the outcome at Time 1, and the importance of both personal and sexual needs. Results supported two of the three hypotheses, indicating that the importance of The Benefits of Valuing Being Friends 19 affiliative needs is positively associated with increases in romantic commitment (β = .15, t(178) = 2.04, p = .043), and sexual need fulfillment (β = .23, t(178) = 3.26, p = .001), but not with increases in love (β = .04, t(178) = 0.54, p = .59). Testing Hypothesis 8b. We further hypothesized that the importance of personal needs and sexual needs will not be significantly associated with positive change in romantic commitment, love, or sexual need fulfillment by Time 2 above and beyond the importance of affiliative needs. Results from multiple regression analysis provided support for this hypothesis. Controlling for the importance of affiliative and sexual needs, personal needs are not significantly associated with increases in commitment (β = -.02, t(178) = -0.25, p = .80), love (β = .05, t(178) = 0.78, p = .44) or sexual need fulfillment (β = .01, t(178) = 0.11, p = .91). Controlling for the importance of affiliative and personal needs, the importance of sexual needs is also not significantly associated with commitment (β = -.06, t(178) = -0.94, p = .35) or love (β = -.06, t(178) = -0.96, p = .34), and whereas it is significantly associated with sexual need fulfillment (β = -.20, t(178) = -2.93, p = .004), the association is negative. Furthermore, this association is negative even when not controlling for the importance of affiliative and personal needs (β = -.19, t(180) = -2.70, p = .008), ruling out the possibility that multicollinearity flipped the sign of this association. Testing Hypothesis 9. Hypothesis 9 held that those who reported at Time 1 that fulfillment of affiliative needs was most important (compared to personal and sexual needs) would be significantly more likely to remain involved by Time 2 than would those who endorsed that either personal or sexual need fulfillment was most important. To test this hypothesis, participants were first categorized into one of four categories based on their endorsement of the subjective importance placed on fulfillment of the specific needs: those who rated the importance The Benefits of Valuing Being Friends 20 of affiliative needs highest (32.9%), those who rated the importance of personal needs highest (15.8%), those who rated the importance of sexual needs highest (21.4%), and those who rated no need as being more important than other needs (29.9%). The fourth group, the participants who endorsed no preference (i.e., they endorsed equivalent preference for their two most important needs, or for all three types of needs), was excluded from analyses (n = 54 out of total sample N = 184). The remaining three groups were analyzed to determine whether they differed in terms of likelihood of dissolution by Time 2. Results from a one-way ANOVA indicated that there was, in fact, a difference in the likelihood of dissolution among the three groups (F(2, 133) = 3.33, p = .039). Examining the least-square means for the three groups indicated that only 11.7% of participants in the group valuing the fulfillment of affiliative needs most had left their relationships by Time 2, whereas 23.7% of those valuing personal need fulfillment most and 20.9% of those valuing sexual need fulfillment most had left their relationships. Thus, the hypothesis that people who value affiliative fulfillment over personal and sexual fulfillment will be more likely to remain in an intact relationship than those who value personal or sexual fulfillment more received support. General Discussion Results from the current studies provide evidence that valuing the friendship aspect of a romance is associated with positive relational outcomes both concurrently and over time. In terms of the cross-sectional benefits of friendship, both studies provided evidence that valuing the friendship component of a romantic relationship is associated with greater concurrent commitment to that relationship, greater experienced love for the partner, and interestingly, greater sexual need fulfillment. Further, results indicated that valuing the friendship is associated with increases in relationship commitment, love, and sexual need fulfillment over time. In line The Benefits of Valuing Being Friends 21 with the notion that friendship within a romantic relationship buffers the partners from dissolution, results indicated that valuing the friendship with the partner is positively associated with romantic relationship persistence over time. In Study 1, we employed a behavioral measure of the value placed on the friendship aspect of the relationship, by measuring participants’ investment in their friendship. We found that those participants who had a greater amount of valued linkages (i.e., both past investment and future plans to invest in their friendship) reaped greater relational benefits. We conceptually replicated this finding in Study 2, where we employed a more direct measure of the value participants placed on the friendship by asking them how much importance they placed on affiliative need fulfillment within their relationship. Again, the amount of value placed on friendship led to romantic relational benefits. What Study 2 provided that was unique from both Study 1 and past research on the importance of friendship to a romance was that all of these relational outcomes were predicted controlling for the effects of the importance placed on sexual need fulfillment and personal need fulfillment. As such, Study 2 provides evidence that placing importance on the friendship in a romance is uniquely associated with these outcomes. Furthermore, in Study 2 we conducted a direct test of whether valuing the friendship more than other aspects of the relationship provides benefits to the relationship. Indeed, we found that those participants who placed greater importance on their friendship than on the personal and sexual aspects of their relationships were less likely to have left their relationship by Time 2 than those participants who placed greater importance on either the personal or sexual aspects of their relationships. There were two instances where our hypotheses did not receive total support, both in Study 2. First, we hypothesized that the importance of sexual need fulfillment would not be The Benefits of Valuing Being Friends 22 associated with the outcomes when controlling for the importance of affiliative and personal need fulfillment. We had extended this hypothesis as a rigorous test of whether valuing the friendship was uniquely associated with the outcomes, providing prediction that valuing other aspects of the relationship could not. In our results, the importance of both affiliative and sexual need importance significantly predicted both concurrent sexual need fulfillment and change in sexual need fulfillment over time. Whereas the association between the importance placed on sexual need fulfillment and sexual need fulfillment was positive when measured concurrently, when measured across time, the association becomes negative. This suggests that in the long term, placing great importance on sexual need fulfillment within a relationship can lead to ironic negative consequences for sexual need fulfillment. Importantly, the importance placed on affiliative needs is associated with positive change in sexual need fulfillment over time, suggesting that valuing the friendship part of a relationship specifically yields benefits for the relationship partners, even with regard to sexual need fulfillment, and that these benefits are unique to valuing the friendship. Second, valuing the friendship was not consistently associated with an increase in love over time when controlling for the importance of sexual and personal need fulfillment. That is, the pattern we obtained in Study 1 was not replicated in Study 2. This led us to consider whether the two measures used to assess valuing the friendship have different implications for change in love over time. Perhaps it is the case that for love to grow, it is necessary to not only place great importance on the fulfillment of affiliative needs, but also necessary to invest in the friendship. Alternatively, it may be the case that the amount of love participants experienced in our Study 2 sample was so great as to leave little room for increase over time. Indeed, the mean of love at Time 1 was quite high (M = 7.6 on a scale from 1 to 9), but it was not higher than the mean for The Benefits of Valuing Being Friends 23 love in Study 1 (M = 7.6), in which we did find friendship investment predicted an increase in love by time 2. This gives us greater confidence that perhaps there is a difference between finding friendship important and investing in it with regard to love, but future research is needed to determine if this is indeed the case, and to understand the mechanism by which this occurs. Strengths and Limitations Together, the results of these two studies provide evidence that valuing friendship in a romance is associated with positive outcomes. The strengths of this work lie in the design of the studies. By collecting data at two time-points in both studies, we were able to show that friendship provides benefits not only concurrently, as past work has shown, but also predicts the romantic relationship getting stronger over time. Additionally, by collecting two different measures for “valuing friendship,” we were able to show that the benefits of valuing friendship are evident both when participants invest in their friendship as well as when they simply place importance on it, but that a difference may exist with regard to love. Perhaps for love to increase over time, it is not enough to only place importance on friendship, but one must also invest in the friendship. Future research is needed to ascertain whether this is true, but by collecting two different measures of valuing the friendship, this nuance was able to emerge. This study is not without limitations, however, with the most notable being the somewhat homogeneous sample. We opted to sample non-marital partners for many reasons, including greater variability in our outcomes of interest as well as a greater likelihood of sampling participants who placed great importance on the personal and sexual aspects of the relationship, but future work is now necessary to determine if these findings generalize to samples in relationships with greater amounts of tangible investments (i.e., married couples, partnered couples). The Benefits of Valuing Being Friends 24 Finally, the current studies suggest that valuing the friendship aspect of a relationship yields benefits for the relationship, but the mechanism behind this effect remains unclear. By examining what aspects of the relationship individuals value, we have begun to consider what outcomes they perceive as desirable and are likely to strive towards. Like when pursuing goals in various domains, pursuing positive friendship-related outcomes is likely to be associated with selectively attending to situations and information that will facilitate successful attainment of these outcomes. In other words, individuals may be particularly attuned to opportunities to strengthen their friendship with their partner, which may lead to greater amounts of relationship promoting behaviors, such as capitalizing on positive events. Additionally, with friendshiprelated outcomes in mind, individuals are likely to engage in transformation of motivation, in which they consider and place the broader interests of their friendship with their partner ahead of immediate, self-interested instincts (Kelley & Thibaut, 1978). This process leads to many positive behaviors, such as giving the partner the benefit of the doubt in ambiguous situations, and being willing to sacrifice, forgive, and accommodate a partner when necessary (Kelley & Thibaut, 1978). Future research examining what exact behaviors and processes arise that facilitate the movement from valuing the friendship to attaining positive outcomes for the relationship would be beneficial for a more complete understanding of this association. Conclusion People vary in how much importance they place on different aspects of romantic relationships, but as friendship is often considered a central aspect of romances, we hypothesized that valuing the friendship may fortify the romance against negative outcomes and serve as a buffer against romantic dissolution. Indeed, results indicated that investing in, and placing importance on the friendship aspect of the relationship were associated with positive concurrent The Benefits of Valuing Being Friends 25 outcomes, as well as positive outcomes over time. Furthermore, the evidence suggests that these benefits come from valuing friendship specifically, rather than any other aspect of the relationship. The Benefits of Valuing Being Friends 26 Footnotes To ensure participants were able to mentally parse the investments they made to the friendship aspect of their relationships from the investments they made to their romance more broadly, we also collected the six valued linkage items worded to apply to the “romantic” relationship (e.g., “I have put a great deal into our romantic relationship that I would lose if the romantic relationship were to end”). A confirmatory factor analysis comparing a one factor model (all items driven by a single underlying latent dimension) to a two-factor model (with one latent factor driving responses to friendship items and a second latent factor driving responses to the romantic items) revealed that the two-factor model fit the data significantly better (χ(1) = 86.75, p < .001). This supports the notion that participants are able to consider the investments they have made and plan to make in the friendship aspect of their relationships uniquely from their investment to the romance broadly. It is possible that the associations found across time were explained by the fact that levels of friendship investment led some participants to terminate their relationships, which in turn led to drops in their levels of commitment, love, and sexual need fulfillment. To rule out this possibility, we ran all analyses controlling not only for the Time 1 measure of the outcome, but also controlling for whether the relationship remained intact by Time 2. Doing so, we found that friendship investment remains positively associated with increases in romantic commitment (β = .12, t(185) = 2.72, p = .007), as well as love (β = .09, t(185) = 1.91, p = .059). Perhaps due to the fact that sexual need fulfillment stems from a dyadic event that stops occurring once a relationship ends, friendship investment did not predict change in sexual need fulfillment over time when controlling for whether the relationship remained intact (β = .06, t(185) = 1.16, p > .20, p = .25). The Benefits of Valuing Being Friends 27 We also ran all analyses again controlling not only for the Time 1 measure of the outcome and the importance of the other needs, but also whether the relationship remained intact by Time 2. Doing so, we found that affiliative need importance remains positively associated with an increase in romantic commitment (β = .08, t(177) = 2.04, p = .042), and a marginal increase in sexual need fulfillment (β = .08, t(178) = 1.81, p = .072). Affiliative need importance did not predict change in love over time when controlling for whether the relationship remained intact (β = -.01, t(178) = -0.15, p = .86), which was expected given that this association was not significant without the added covariate. The Benefits of Valuing Being Friends 28 ReferencesAgnew, C. R., Arriaga, X. B., & Goodfriend, W. (2006). On the (mis)measurement of premaritalrelationship breakup. Paper presented at the International Association for RelationshipResearch.Agnew, C. R., Arriaga, X. B., & Wilson, J. E. (2008). Committed to what? Using the Bases ofRelational Commitment Model to understand continuity and changes in socialrelationships. In J. P. Forgas & J. Fitness (Eds.), Social relationships: Cognitive, affectiveand motivational processes (pp. 147-164). New York: Psychology Press.Aron, A., & Westbay, L. (1996). Dimensions of the prototype of love. Journal of Personalityand Social Psychology, 70, 535-551.Berndt, T. J. (2002). Friendship quality and social development. Current Directions inPsychological Science, 11, 7-10.Fehr, B. (1988). Prototype analysis of the concepts of love and commitment. Journal ofPersonality and Social Psychology, 55, 557-579.Fehr, B. (1994). Prototype-based assessment of laypeople's views of love. PersonalRelationships, 1, 309-331.Fehr, B. (1996). Friendship processes. Thousand Oaks, CA: Sage.Fehr, B., & Russell, J. A. (1991). The concept of love viewed from a prototype perspective.Journal of Personality and Social Psychology, 60, 425-438.Fine, M. A., & Harvey, J. H. (2006). Handbook of Divorce and Relationship Dissolution. London: Lawrence Earlbaum Associates.Fletcher, G. J. O., & Kininmonth, L. (1992). Measuring relationship beliefs: An individualdifference measure. Journal of Research in Personality, 26, 371-397. The Benefits of Valuing Being Friends 29 Fletcher, G. J. O., Rosanowski, J., & Fitness, J. (1994). Automatic processing in intimatecontexts: The role of close-relationship beliefs. Journal of Personality and SocialPsychology, 67, 888-897.Goodfriend, W., & Agnew, C. R. (2008). Sunken costs and desired plans: Examining differenttypes of investments in close relationships. Personality and Social Psychology Bulletin,34, 1639-1652.Grote, N. K., & Frieze, I. H. (1994). The measurement of friendship-based love in intimaterelationships. Personal Relationships, 1, 275-300.Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence.New York: Wiley.Kim, J., & Hatfield, E. (2004). Love types and subjective well-being: A cross cultural study.Social Behavior and Personality, 32, 173-182.Le, B., Smoak, N. D., Agnew, C. R., Korn, M. S., & Mutso, A. A. (2010). Predicting non-maritalromantic relationship dissolution: A meta-analytic synthesis. Personal Relationships, 17,377-390.Regan, P. C., Kocan, E. R., & Whitlock, T. (1998). Ain't love grant! A prototype analysis of theconcept of romantic love. Journal of Social and Personal Relationships, 15, 411-420.Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of theinvestment model. Journal of Experimental Social Psychology, 16, 172-186.Rusbult, C. E., Martz, J. M., & Agnew, C. R. (1998). The Investment Model Scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size.Personal Relationships, 5, 357-391. The Benefits of Valuing Being Friends 30 Sanderson, C. A., Keiter, E. J., Miles, M. G., & Yopyk, D. J. A. (2007). The association betweenintimacy goals and plans for initiating dating relationships. Personal Relationships, 14,225-243.VanderDrift, L. E., Agnew, C. R., & Wilson, J. E. (2009). Non-marital romantic relationshipcommitment and leave behavior: The mediating role of dissolution consideration.Personality and Social Psychology Bulletin, 35, 1220-1232. The Benefits of Valuing Being Friends 31 Table 1Principal Component Analysis Loadings (Study 2) Affiliative Needs Personal Needs Sexual Needs NeedSecurity0.83-0.030.12Care-giving0.82-0.010.10Companionship0.650.07-0.29Independence-0.370.860.05Self-Improvement0.260.76-0.04Self-Expansion0.310.700.02Sexual Contact0.030.030.96 Inter-factor CorrelationsAffiliative Needs-Personal Needs0.23-Sexual Needs0.050.04-Note. Bolded values represent the component on which the items were determined to load.