Journal of Happiness Studies

, Volume 11, Issue 3, pp 293–313

Close Relationships and Happiness Among Emerging Adults

Authors

Research Paper

DOI: 10.1007/s10902-009-9141-x

Cite this article as:
Demir, M. J Happiness Stud (2010) 11: 293. doi:10.1007/s10902-009-9141-x

Abstract

The present investigation examined the role of multiple close relationships (mother, father, best friend, and romantic partner, if any) in happiness among emerging adults with and without a romantic partner. The results for those without a partner (n = 152) revealed that only the relationship experiences with mother and best friend were predictive of happiness. On the other hand, the findings for those with a partner (n = 159) showed that only three factors, namely mother–child relationship quality, romantic relationship quality and conflict were predictive of happiness. The results for this group also suggested that romantic relationship quality was protective of best friendship conflict; moreover, best friendship quality did not buffer the negative impact of romantic partner conflict on happiness, suggesting a less important role of best friends in happiness. In other words, the findings suggest that when emerging adults are involved in a romantic relationship, friends’ importance in happiness might be less pronounced or not pronounced at all. The results were discussed in light of the literature and suggestions were made for future research.

Keywords

Close relationshipsCross-domain bufferingEmerging adulthoodHappinessRelationship quality

1 Introduction

Individuals enjoy a variety of close relationships in their lives and the quality of these relationships has been associated with happiness (Lucas and Dyrenforth 2006). Indeed, the role of close relationships in happiness has been called as the “deep truth” (Myers 1992). This argument and similar others (e.g., Argyle 2001) are based on studies that investigated one particular relationship (e.g., friendship) without taking other close relationships into account. It could be that the role of different relationships in one’s happiness might change when individuals experience normative developmental changes in their lives (Levitt 1991). For example, do the role close relationships play in happiness change when one is involved in a romantic relationship during emerging adulthood? This is a theoretically essential question because during this phase of life, most individuals, if not all, strive to establish and maintain romantic relationships (Arnett 2000). At the same time, it is critical to note that not every emerging adult is involved in a committed romantic relationship and some people might prefer to be single (DePaulo and Morris 2005). Accordingly, the present study gathered relationship quality and conflict information from multiple figures [mothers, fathers, best friends and romantic partners (if any)] and investigated their role in the happiness among emerging adults with and without a romantic partner. The reason for focusing on these relationships stems from the fact that researchers consider family relationships, friendships and romantic relationships (dating and marriage) as the most important close relationships in one’s life (Clark and Graham 2005).

Investigation of multiple relationships simultaneously is important at least for two reasons. First, this practice allows one to test the predictive ability of different relationships in different life conditions. For example, it could be that friendship quality does not contribute to well-being when considered together with family and romantic relationships (e.g., Whisman et al. 2000). Second, gathering relationship quality and conflict information from multiple relationships allows one to test for relationship specific and cross-domain buffering interactions (e.g., quality of friendship buffering the negative effects of conflict with romantic partner). Overall, the simultaneous investigation of multiple close relationships has the potential to promote our understanding of the role of close relationships in happiness.

It is essential to describe happiness before the relevant literature is presented. Happiness, or subjective well-being, as it is interchangeably used in the literature, refers to the cognitive and affective evaluations of one’s own life and is defined in terms of global life satisfaction, presence of positive affect and absence of negative affect (Diener 1984, 1994). These three components are different constructs and might require further research to understand each of them separately (Diener et al. 1999). However, the three components are substantially correlated and research suggests a higher order factor (Diener 1994; Kasser and Sheldon 2002; Sheldon et al. 2005), which is referred to as happiness.

1.1 Relationships with Parents, Friends, Romantic Partners and Happiness

Emerging adults are likely to enjoy several different relationships. One commonality in these different relationships is the features experienced. Theory suggests that there are two main dimensions of close relationships: relationship quality and conflict (Weiss 1974; Cutrona and Russell 1987). The former refers to features such as help and intimacy experienced in the relationship, whereas the latter refers to the frequency of conflict experienced within the relationship (Furman and Buhrmester 1985). In the present study, these two main dimensions are assessed across different relationships. What follows is a literature review describing the association between close relationships and happiness.

The quality of relationships with parents plays an essential role in the well- emerging adults (Aquilino 1997; Prager 1995). Supporting the theoretical arguments, both cross-sectional and longitudinal studies show that relationship quality and conflict with family members are associated with happiness and general well-being among emerging adults (Amato 1994; Wel et al. 2000). It is important to note that some studies document mother–child relationships as having a stronger influence on well-being (e.g., Field et al. 1995) whereas others report stronger influence for father–child relationships (e.g., Rohner and Veneziano 2001). The take home message, though, is that relationships with mothers and fathers both influence the well-being of emerging adults.

Friends constitute an important part of many people’s lives across the life-span (Hartup and Stevens 1997). Theory suggests that friendships are an important source of well-being (Argyle 2001; Baumeister and Leary 1995). In support of theory, decades of empirical research documents that friendship quality is related to happiness among adults (Demir and Weitekamp 2007; Diener and Seligman 2002; Lyubomirsky et al. 2006). As for friendship conflict, Demir and his colleagues (Demir et al. 2007; Demir and Weitekamp 2007) and Rathur (2004) documented a negative association between the two among emerging adults. Thus, substantial amount of research clearly documents a relationship between friendship experiences and happiness.

Formation and maintenance of romantic relationships are central to the lives of emerging adults (Arnett 2000). Supporting this notion are findings that romantic relationships constitute an important part of the everyday lives of college students (Gable et al. 2004) and most college students consider their romantic involvements as the closest relationships they have (Berscheid et al. 1989). In support of the theoretical arguments regarding the importance of romantic relationships in well-being (Berscheid and Regan 2005; Reis 2001), a considerable amount of research document has documented an association between romantic relationships (involvement in and/or quality-conflict of) and happiness across the life-span (Campbell et al. 2005; Kamp Dush and Amato 2005; Berry and Willingham 1997; see Keyes and Waterman 2003 for a review). Overall, there is substantial evidence showing that the quality and conflict of romantic relationships are related to happiness.

As it should be clear by now, the quality of close relationships is related to happiness when studied in isolation. On the other hand, a different picture emerges when multiple relationships are investigated simultaneously. When researchers gather relationship quality information from multiple figures (family members, romantic partners and friends); friendship quality either does not contribute to well-being or makes the lowest contribution. For example, Walen and Lachman (2000) gathered interview and self-report data from more than 3,000 American adults who were married and cohabiting. Information about social support and strain experienced in family relationships, friendships and romantic relationships were obtained from the participants. They reported that support received from and strain experienced with friends contributed to mental and psychological well-being less than the relationship experiences with family members and partners. Similar findings were obtained in other nationally representative samples as well (Bertera 2005). Other studies reported that friendships experiences do not contribute to well-being when other relationships are taken into account. For example, Whisman et al. (2000) gathered interview data from 5,000 married adults across the life-span and assessed psychological well-being as well as relationship experiences with romantic partners, friends and relatives (e.g., family members). They reported that friendship experiences were not related to psychological well-being when controlling for other social relationships. Okun and Keith (1998) reported similar findings in a nationally representative sample as well. In another nationally representative sample (n > 2,000), Taylor et al. (2001) gathered social support and network data from single, married and divorced African Americans via interviews. They reported that only the number of friends was related to well-being, however, its role was negligible as compared to the role of family and partner relationships .Overall, available research suggests that the role of friendship quality in well-being might be less pronounced among those involved in a romantic relationship. This point necessitates to consider why this might be the case and how different it might be for emerging adults not involved in a romantic relationship.

1.2 Emerging Adults With and Without a Romantic Partner

As several theorists have argued, the development of close and intimate romantic relationships with others is a critical task of the transition period from adolescence to adulthood (Arnett 2000; Erikson 1982). Supporting the theoretical arguments, recent research documented that emerging adults perceive finding a romantic partner and developing intimate relationships with another person as very important (Arnett 2006). Emerging adults are searching for a lifelong partner and this has the potential to shape not only how love relationships but also other close relationships are experienced (e.g., friendships). Role theory considers this point and suggests that new roles acquired in life (e.g., becoming a romantic partner) may influence the way one organizes his/her social life (Duvall 1971; see Carbery and Buhrmester (1998) and Collins and van Dulmen (2006) for similar arguments). Accordingly, individuals are likely to reorganize the hierarchy of their close relationships.

Social convoy model suggests that one’s relationships are hierarchically organized and are likely to change when one experiences normative developmental changes (Levitt 1991; Levitt et al. 1986). That is, even though one might have a network of relationships that continues for a lifetime, normative life transitions (e.g., developing a romantic relationship) might change the relationship hierarchy and which relationship makes the biggest contribution to one’s well-being (Levitt et al. 1986, 1993a, b; Ruehlman and Wolchik 1988). Accordingly, it might be normative to place romantic relationships at the top of the hierarchy of close relationships (Clark and Graham 2005) during emerging adulthood, especially when one is involved in a romantic relationship. This suggests that close friendships might lose their significance when attention is directed to romantic relationships (Collins and Laursen 2000). Indeed, theory and research support this contention.

According to the dyadic withdrawal hypothesis (Johnson and Leslie 1982), as individuals get involved in one relationship, they will decrease their involvement in other relationships. In the case of developing romantic relationships, research has shown that individuals have fewer contacts and spend less time with their friends (Johnson and Leslie 1982; Fischer et al. 1989) and report a decrease in friendship networks (Milardo 1982; Kalmijn 2003). Additional research on the topic further suggests that one not only spends less time with his/her friends but also relies less on them to satisfy relationship provisions (e.g., intimacy; Reis et al. 1993), which is consistent with the idea of changing relationship hierarchies. That is, friends might be the main providers of social provisions among those who are single, but they fall behind romantic partners when one is involved in a romantic relationship (Carbery and Buhrmester 1998; Furman and Buhrmester 1992).

Theory and research suggest that when one is facing a normative developmental event (establishing romantic relationships), the role of friendships in one’s life and well-being might go through some changes. Accordingly, it seems reasonable to propose that among emerging adults involved in a romantic relationship, friendship quality might not have a strong influence on happiness as romantic partners do.

The preceding review of theory and research might give the reader the idea that all emerging adults are involved in a romantic relationship and go through normative changes as predicted by theory. Even though majority of emerging adults might eventually establish romantic relationships and get married (Arnett 2000; Prager 1995), it is possible to find emerging adults not involved in romantic relationships either by preference or lack of opportunity (DePaulo and Morris 2005). Supporting these arguments are studies (e.g., romantic attachment) documenting that between 26 and 50% of any emerging adult sample is not involved in a romantic relationship (Collins 1996; Doherty and Feeney 2004) and the number of singles (without a committed romantic relationship) in the US increasing (DePaulo and Morris 2005). This suggests that relationships other than romantic ones might play an important role in the happiness of single emerging adults. Accordingly, the aims of the present investigation offered an excellent opportunity to examine the role of multiple close relationships among those with and without a romantic partner.

1.3 Buffering Effects of Relationship Quality and Conflict

Gathering information on relationship quality and conflict from multiple figures raises the issue of buffering effect. Social support research has documented that supportive exchanges could offset the detrimental impact of negative exchanges on well-being (Cohen and Wills 1985). There are two types of buffering reported in the literature: relationship specific and cross-domain buffering. The former is the classically studied buffering effect where the support received from someone (e.g., mother) offsets the negative impact of conflict experienced with the same person. For example, this type of buffering effect would suggest that high levels of friendship quality might offset the potential negative impact of the conflict experienced with the friend. The latter is investigated when multiple relationships are studied simultaneously and is observed when a supportive relationship with someone (e.g., romantic partner) buffers the negative influence of conflict experienced with others (e.g., friendship).

Empirical studies have found support for both types of buffering (Lepore 1992; Walen and Lachman 2000; see Okun and Keith 1998 for a review of significant and null findings). For example, Walen and Lachman (2000) investigated buffering effects in a nationally representative sample consisting of married and cohabiting adults by gathering data from the romantic partner, family and the friends. They reported that support from friends buffered the negative impact of romantic partner conflict on well-being. That is, adults experiencing supportive friendships were less likely to be influenced by conflicts in their romantic relationships. However, considering the fact that previous research findings showing support for buffering effects was not consistent across studies (Okun and Keith 1998) and the fact that the present investigation focused on broader relationship quality but not social support, no specific hypotheses were formulated regarding relationship-specific and cross-domain buffering effects. In other words, analyzes pertaining to buffering effects were exploratory in nature.

2 Hypotheses

In the light of theory and the literature reviewed, two specific hypotheses were developed.
  1. (a)

    Relationship quality would be positively correlated with happiness across all relationships among emerging adults with and without a romantic relationship.

     
  2. (b)

    Relationship conflict would be negatively correlated with happiness across all relationships among emerging adults with and without a romantic relationship.

     
  3. (c)

    When considered with other relationships, friendship quality would be a significant predictor of happiness among emerging adults without a romantic partner whereas it would not be an important predictor among those involved in a romantic relationship.

     

3 Method

3.1 Participants

The original sample consisted of 314 (187 women, 124 men) young adults attending a Midwestern university. Three participants were eliminated because they were over 30; this needed to be done in order to have a sample within the age range of emerging adulthood. Of the original sample, 51% (n = 159) was currently involved in a romantic relationship. This rate is comparable to other studies (e.g., Doherty and Feeney 2004).

The sample making up those involved in romantic relationships consisted of 159 (92 women, 67 men) emerging adults. The age of this group ranged from 18 to 28 (M = 22.75, SD = 4.74). The ethnic distribution of the sample was as follows: 39% Caucasian and African American each (n = 62); 7% Asian (n = 11); and 15% other (n = 24). Participants were involved in romantic relationships with a duration of 26.62 months (SD = 22.46). The average length of the romantic relationship was 33.73 months (SD = 31.65) with a range from 2 to 89 months.

The sample for those not involved in a romantic relationship consisted of 152 (95 women; 57 men) emerging adults. The age of this group ranged from 18 to 29 (M = 20.50, SD = 2.55). The ethnic distribution of the sample was as follows: 38% Caucasian and African American, each (n = 57); 7% Asian (n = 10); and 17% other (n = 28). It is important to note that the ethnic and gender composition of the samples across the two groups were in line with the student population of the university.

3.2 Procedure

A psychology student pool was used to recruit participants. Announcements were made in classrooms and flyers were posted in the psychology department. Those who wanted to participate in the study either took the survey with them to complete on their own time or completed the questionnaire packet in our lab. The packet included a consent form, a basic demographic information sheet and a battery of questionnaires. To ensure privacy, participants were given envelopes to enclose the completed surveys. Those taking the surveys with them placed the envelopes in a designated location or turned them in directly to the researcher the next day. Completion of the survey lasted approximately forty minutes and participants earned extra credit for their psychology classes.

In the final sample, 216 participants completed the survey at our lab and 95 participants took the survey with them and returned it the next day. Comparison between the two groups (lab vs. home) revealed no significant differences on any of the variables investigated.

3.3 Measures

3.3.1 Relationship Quality and Conflict

A shortened version of the network of relationship inventory (NRI) was used to assess relationship quality and conflict (Furman and Buhrmester 1985). The scale has been used in previous research to make comparisons across relationships (Furman and Buhrmester 1992) and it has been used among adults as well (e.g., Bagwell et al. 2005). In the present study, the participants were asked to rate their relationships with their mothers (or mother like figure), fathers (or father like figure), best friends and romantic partners.

The shortened version used to measure relationship quality assessed four provisions within each relationship: companionship, intimacy, reliable alliance and affection. Each relationship provision was assessed with three items. Sample items include “How often do you go places and do enjoyable things with this person?” and “How much do you share your secrets and private feelings with this person?” Respondents were asked to rate how much each feature occurred in their relationship on a five-point scale ranging from 1 (little or none) to 5 (the most). A sum of these 12 items was used to create the relationship quality composite score for the relationships rated, higher scores indicating higher relationship quality. The internal consistency of the scales across relationships is reported in Table 1.
Table 1

Internal consistency of the scales used in the study

 

With a romantic partner

Without a romantic partner

Mother quality

.89

.90

Mother conflict

.91

.87

Father quality

.92

.93

Father conflict

.90

.92

Friend quality

.87

.92

Friend conflict

.90

.85

Romantic quality

.91

Romantic conflict

.94

Life satisfaction

.88

.90

Positive affect

.86

.84

Negative affect

.85

.87

Three-item version of the conflict subscale of NRI was used to assess the amount of conflict experienced in the relationships. Sample item includes ‘‘How much do you and this person get upset with or mad at each other?’’ The rating scale was the same with the relationship quality scale. The three items were summed to create a relationship conflict composite score for the relationships rated, higher scores indicating higher relationship conflict.

As for validity, Furman and Buhrmester (1985, 1992) reported that the scale was sensitive to gender differences (women scoring higher than men) and developmental changes experienced in close relationships from childhood to emerging adulthood. It was also documented that the scale was related to well-being (e.g., happiness) (Bagwell et al. 2005; Demir et al. 2007; Hussong 2000). Research showed good to moderate reliabilities for the subscales and the overall scale (Furman and Buhrmester 1992; Kuttler and La Greca 2004; Demir et al. 2007). Please see Table 1 for the internal consistency of the scales obtained in the present study across different relationships among those with and without a romantic partner.

3.3.2 Happiness

Happiness was assessed with the satisfaction with life scale (SWLS) (Diener et al. 1985) and the positive and negative affect schedule (PANAS) (Watson et al. 1988). The SWLS assesses the global cognitive evaluations of one’s life. The scale consisted of five items and respondents were asked to rate their agreement with the items on a seven-point scale ranging from (1) strongly disagree to (7) strongly agree. Sample item includes “If I could live my life over, I would change almost nothing”. Research supporting the validity of SWLS showed that certain groups reported the lowest levels of life satisfaction (e.g., psychiatric patients, abused women) compared to normal groups and college students (Pavot and Diener 1993). SWLS was also shown to be positively related to an array of positive well-being measures (Lyubomirsky and Lepper 1999; Pavot et al. 1991) and to be negatively associated with distress and depression (Arrindell et al. 1991). As for reliability, Diener et al. (1985) and others (Demir and Weitekamp 2007; Pavot et al. 1991) reported moderate to strong internal consistency and test-retest reliability for the scale.

The PANAS was used to assess general positive and negative affect. The PANAS consisted of ten mood states for positive affect (PA) (e.g., enthusiastic) and ten for negative affect (NA) (e.g., nervous). Respondents were asked to rate, in general, the extent to which they feel each mood on a 1 (very slightly or not all) to 5 (extremely) scale. Supporting the validity of PANAS, both positive and negative affect subscales were related to other scales assessing mood, distress and well-being in the expected directions (Crawford and Henry 2004; Lyubomirsky and Lepper 1999; Watson et al. 1988). As for reliability, Watson et al. (1988) and others (Crawford and Henry 2004; Kasser and Sheldon 2002) reported moderate internal consistency and test-retest reliability for the scales (please see Table 1 for the internal consistency of the happiness measures used in this study).

In the present study, correlations between these constructs ranged from −.24 to .39. Also, a principal components analysis of the three well-being measures (after recoding negative mood) revealed a single factor, which accounted for 47% of the variance. Accordingly, an aggregate happiness score was formed by standardizing and summing positive mood and life-satisfaction, then subtracting negative mood (Diener 1994). It is important to note that creation of aggregate score for happiness (using the same measures and exact procedures) has been reported by others as well (e.g., Kasser and Sheldon 2002).

4 Results

The correlations between the study variables for those with and without a romantic partner are reported in Tables 2 and 3. Two points are noteworthy. First, in support of the hypothesis (a), the quality of every relationship was related to happiness in the expected directions in both groups (with/out a romantic partner). Second, standing in contrast to the hypothesis (b), were the findings that relationship conflict with the mother and father was not related to happiness in either group. On the other hand, even though friendship conflict was negatively associated with happiness among those without a romantic partner, it was not related to happiness among those in a romantic relationship.
Table 2

Correlations among the study variables for those not involved in a romantic relationship (n = 152)

 

1

2

3

4

5

6

7

8

9

10

11

1. Gender

          

2. Mother quality

.24**

         

3. Mother conflict

.10

−.09

        

4. Father quality

.10

.38**

.06

       

5. Father conflict

−.01

−.07

.22**

.06

      

6. Friend quality

.20*

.17*

.16*

−.01

.13

     

7. Friend conflict

−.28**

−.13

−.04

.07

.36**

−.04

    

8. Life satisfaction

.19*

.31**

−.02

.27**

−.04

.29**

−.18*

   

9. Positive affect

−.01

.01

−.09

.08

.02

.02

.02

.38**

  

10. Negative affect

.05

−.15

−.06

−.13

.16*

−.15

.29**

−.34**

.07

 

11. Happiness

.06

.24**

−.02

.22**

−.07

.25**

−.23**

.82**

.62**

−.62**

M

3.74

2.48

3.14

2.25

3.93

1.91

4.59

3.22

2.51

.08

SD

.71

1.12

.89

1.17

.96

.91

1.30

.45

.54

1.90

P < .05; ** P < .01

Table 3

Correlations among the study variables for those involved in a romantic relationship (n = 159)

 

1

2

3

4

5

6

7

8

9

10

11

12

13

1. Gender

            

2. Mother quality

.12

           

3. Mother conflict

.04

.05

          

4. Father quality

.01

.58**

.04

         

5. Father conflict

.03

.17*

.26**

.18*

        

6. Friend quality

.00

.15*

.04

.14

.01

       

7. Friend conflict

−.02

.07

.24**

−.01

.14

−.13

      

8. Romantic quality

.12

.30**

.07

.30**

.02

.22**

−.07

     

9. Romantic conflict

−.01

−.12

.24**

−.09

.13

−.16*

.19*

−.06

    

10. Life satisfaction

−.04

.31**

−.08

.31**

−.15

.19*

−.15*

.35**

−.12

   

11. Positive affect

.08

.24**

.02

.18*

.09

.24**

.03

.20*

.11

.32**

  

12. Negative affect

.26**

.05

.24**

−.04

.18*

−.02

.23**

−.09

.32**

−.20**

−.30**

 

13. Happiness

−.12

.26**

−.13

.28**

−.13

.25**

−.18*

.34**

−.19*

.82**

.54**

−.49**

M

3.50

2.15

2.94

2.30

3.85

1.62

3.91

1.60

4.71

3.30

2.61

.15

SD

.75

1.01

.90

.87

.91

1.17

.69

.85

1.41

.45

.55

1.81

P < .05; ** P < .01

4.1 Regression Analyzes

Two sets of multiple hierarchical regression analyzes were computed in predicting happiness across the two groups (with and without a romantic partner). Following the recommendations of Aiken and West (1991), variables were centered before the analyzes. In each regression analysis, gender was entered in the first step as a control variable since it was associated with the predictor and/or outcome variables (see Tables 2, 3). The second step involved the respective relationship variables (e.g., mother quality). In the third step, the relationship specific buffering and (e.g., mother quality × mother conflict) cross-domain buffering interactions (e.g., mother quality × friendship conflict) were entered. In the regression among those without a romantic partner, there were three relationship specific buffering interactions (e.g., mother quality × mother conflict) and six cross-domain buffering interactions (e.g., mother quality × friendship conflict). In the regression among those with a romantic partner, there were four relationship buffering interactions and nine cross-domain buffering interactions (e.g., romantic quality × friendship conflict). Considering the fact that no hypotheses were formulated for buffering effects, we set the alpha to .001 for significant interaction effects. In the case of a significant interaction, we computed an additional regression in which the variables making up the interaction were entered in the first step and the interaction term in the second step. To aid in the interpretation of significant interaction(s), effects were plotted and simple slopes were tested to evaluate if they differed significantly from zero (Aiken and West 1991).

4.2 Emerging Adults Not Involved in a Romantic Relationship

Results of the regression analyzes are reported in Table 4. As seen in the table, gender was not a significant predictor of happiness, F (1, 150) = .530, NS. Second step was significant, F (6, 144) = 5.030, P < .01, and accounted for 17% of the variance in happiness. In this step, mother relationship quality, friendship quality and friendship conflict emerged as significant predictors. The final step involving interactions was not significant, F (9, 135) = 1.054, NS and is not reported in the table. Supporting the hypothesis (c), friendship quality and conflict emerged as significant predictors along with mother quality when predicting happiness among those not involved in a romantic relationship.
Table 4

Hierarchical multiple regression predicting happiness among those not involved in a romantic relationship

 

β

Step 1

    Gender (1 = male, 2 = female)

−.09

 

R2 = .01

Step 2

    Mother quality

.18*

    Mother conflict

−.11

    Father quality

.07

    Father conflict

−.05

    Friend quality

.26*

    Friend conflict

−.22*

 

ΔR2 = .17**

β weights are for the final model

P < .05; ** P < .01

4.3 Emerging Adults Involved in a Romantic Relationship

Results of the analyzes are reported in Table 5. First step of the regression involving gender was not significant, F (1, 155) = 1.389, NS. Second step involving the relationship variables were significant, F (8, 147) = 7.181, P < .01, and accounted for 28% of the variance in happiness. As seen from the table, in this step only the mother relationship quality, romantic relationship quality and conflict were significant predictors of happiness. Partially supporting the hypothesis (c), friendship variables did not emerge as significant predictors of happiness. The third step involving the interactions was also significant, F (16, 131) = 2.771, P < .05, and explained an additional 13% of the variance in happiness. In this step, the cross-domain buffering interactions of romantic relationship quality × friendship conflict and friendship quality × romantic relationship conflict were significant. These significant interactions are examined below.
Table 5

Hierarchical multiple regression predicting happiness among those involved in a romantic relationship

 

β

Step 1

    Gender (1 = male, 2 = female)

−.12

 

R2 = .01

Step 2

    Mother quality

.26*

    Mother conflict

−.12

    Father quality

.15

    Father conflict

−.11

    Romantic quality

.20*

    Romantic conflict

−.19*

    Friend quality

−.03

    Friend conflict

−.04

 

ΔR2 = .28**

Step 3

    Romantic quality × friend conflict

.21*

    Friend quality × romantic conflict

−.24**

 

ΔR2 = .13**

β weights are for the final model

P < .05; ** P < .01

As for the first cross-domain buffering interaction (romantic quality × friend conflict), the first step of the regression, F (2, 154) = 12.356, P < .01, and the second step involving the interaction, F (1, 153) = 3.464, P < .05, were significant. The beta value for the interaction was .19. As can be seen in Fig. 1, individuals reporting high levels of friendship conflict experienced lower levels of happiness when their romantic relationship quality was low in quality, t (152) = −2.069, P < .05. On the other hand, there was not a linear relationship between happiness and friendship conflict at high levels of romantic relationship quality, t (152) = .471, NS. In other words, high levels of romantic relationship quality protected the individual from experiencing lower levels of happiness at differing levels of friendship conflict. Overall, the results suggest that romantic relationship quality buffered the negative impact of friendship conflict on happiness.
https://static-content.springer.com/image/art%3A10.1007%2Fs10902-009-9141-x/MediaObjects/10902_2009_9141_Fig1_HTML.gif
Fig. 1

Interaction between romantic relationship quality and best friendship conflict on happiness

As for the other cross-domain buffering interaction (romantic conflict × friendship quality), the first step of the regression, F (2, 154) = 7.991, P < .01, and the second step involving the interaction, F (1, 154) = 7.004, P < .01, were significant. The beta value for the interaction was −.21. As seen in Fig. 2, higher levels of friendship quality were protective of happiness only at low levels of romantic relationship conflict and it was related to lower levels of happiness at high levels of romantic relationship conflict, t (152) = −3.48, P < .01. In other words, cross-domain buffering effect for friendship quality was not observed. On the other hand, there was not a linear relationship between happiness and romantic relationship conflict at low levels of friendship quality, t (152) = −.018, NS. Findings suggest that friendship quality did not buffer the impact of romantic relationship conflict on happiness. Even though this was the case, it is important to note that among those with low levels of friendship quality, levels of romantic relationship conflict were not associated with happiness. Instead, happiness was relatively low regardless of the level of conflict in the romantic relationship.
https://static-content.springer.com/image/art%3A10.1007%2Fs10902-009-9141-x/MediaObjects/10902_2009_9141_Fig2_HTML.gif
Fig. 2

Interaction between best friendship relationship quality and romantic relationship conflict on happiness

5 Discussion

The present study investigated the role of close relationships in the happiness of emerging adults with and without a romantic partner. Simultaneous consideration of multiple close relationships revealed important and interesting findings that advance our understanding of the link between close relationships and happiness. These findings are discussed below.

5.1 Close Relationships and Happiness

To start with, the quality of all relationships and conflict with best friends and romantic relationships were related to happiness at the bivariate level. These findings are in line with extant research documenting the importance of close relationships in the happiness of emerging adults (Amato 1994; Demir and Weitekamp 2007; Wel et al. 2000). It is interesting to note that only conflict experienced with best friends and romantic partners was related to happiness at the bivariate level. A closer look at Tables 2 and 3 suggests that conflict experienced with friends and romantic partners across the two groups (with and without a romantic involvement) was lower in frequency when compared to conflict experienced with mothers and fathers. These findings could be taken to suggest that, despite the lower frequency of conflict, negative relationship experiences with these two figures might become more important during emerging adulthood. This interpretation is consistent with the literature suggesting that best friends and romantic partners constitute an essential aspect of the lives of emerging college students and these two figures might occupy the top places in one’s hierarchy of close relationships (Berscheid et al. 1989; Furman and Buhrmester 1992; Gable et al. 2004; Levitt 1991).

5.2 Happiness Among Emerging Adults Without a Romantic Partner

Even though all relationships were related to happiness to some extent, a different picture emerged when close relationships were considered simultaneously. Specifically, among those not involved in a romantic relationship, only mother and best friend relationship quality and friendship conflict were significant predictors of happiness whereas relationship experiences with fathers did not contribute to happiness. These findings are consistent with prior research documenting the importance of friendship among college students (Shaver et al. 1985). It seems that relationship experiences with best friends, when compared to parents, are likely to be the main source of happiness among those not involved in a romantic relationship.

5.3 Happiness Among Emerging Adults Involved in a Romantic Relationship

The results pertaining to the role of close relationships among those involved in a romantic relationship revealed that only mother and romantic partner relationship quality and romantic partner conflict were significant predictors of happiness. Father relationship quality and conflict did not emerge as a significant predictor, findings consistent with the analyzes reported for those without a romantic partner. Friendship quality and conflict, which were important predictors among emerging adults without a romantic partner, no longer contributed to the happiness of those involved in a romantic relationship. These findings are consistent with prior research (Bertera 2005; Walen and Lachman 2000; Okun and Keith 1998; Whisman et al. 2000) and were qualified by two cross-domain buffering interactions between romantic relationships and friendships. In the first one, romantic relationship quality buffered the negative impact of friendship conflict on happiness. Specifically, the emerging adult was likely to be happy as long as he/she experienced higher levels of romantic relationship quality regardless of the degree of friendship conflict. As for the second interaction, which was not consistent with the typical cross-domain buffering, high levels of friendship quality did not prevent the emerging adult from experiencing lower levels of happiness when romantic relationship conflict was high. Rather, it was related to happiness only when conflict with the romantic partner was low. Overall, the interactions are consistent with the main effect findings and suggest that romantic relationships as compared to friendships play a more important role in the happiness of emerging adults. Even though the interactions lend support to the main findings, they should be interpreted with caution since the analyzes pertaining to interactions were exploratory in nature. Replications of these findings with different samples are necessary to establish confidence in the findings and the interactions reported.

How could one explain the findings? As elucidated earlier, the development of close and intimate romantic relationships with others is a critical task of emerging adulthood (Arnett 2000, 2006) which eventually influences the way one organizes his/her hierarchy of their close relationships (Collins and van Dulmen 2006; Duvall 1971). Consistent with the theory (Levitt 1991; Clark and Graham 2005) and empirical research (Levitt et al. 1993a, b; Reis et al. 1993; Ruehlman and Wolchik 1988), it seems that individuals involved in a romantic relationship might place their friends in a lower position and their romantic partner in the top of their hierarchy of close relationships. As such, romantic partners might become the main providers of social provisions sought in close relationships and main source of happiness when compared to close friends and/or other close relationships (Carbery and Buhrmester 1998; Ruehlman and Wolchik 1988). These arguments also make sense when considering the interactions explained above. For example, positive relationship experiences with the theoretically less important figure in one’s hierarchy of close relationships (e.g., best friend) would not buffer the negative impact of conflict experienced with most important figure in one’s life (Berscheid and Regan 2005).

Even though the findings reported above are unique, one essential point to keep in mind is the possibility that emerging adults might experience a break-up with their romantic partners. The question, then, is what happens to the friendships of emerging adults experiencing a romantic dissolution? According to research, emerging adults experience sadness, anger, distress and depression following the dissolution of their romantic relationships and return to their friends to discuss the break-up (Perilloux and Buss 2008; Sprecher et al. 1998; Sbarra and Emery 2005; Sbarra and Ferrer 2006). Emerging adults also report that they enjoy improved friendships after the break-up and they recover from the emotional turmoil of the romantic dissolution if they receive support from their friends (Frazier and Cook 1993; Tashiro and Frazier 2003). One could interpret this as evidence suggesting the possibility that friends reclaim their importance in one’s network following the dissolution of a romantic relationship. It would be interesting to examine if best friends become the main source of happiness among emerging adults who experienced a romantic break-up.

What do these findings suggest? We believe there are three points that need to be highlighted. First of all, a considerable amount of research (Reis et al. 2000) and theoretical writing (Hinde 1997) points to the importance of friendship in the well-being of emerging adults. We modify these arguments by suggesting that friends’ importance in well-being would be less pronounced or not at all among those with a romantic partner, who are exploring or achieving a normative developmental task, especially when multiple close relationships are considered simultaneously. This argument is in line with research documenting that friends either do not contribute to or take a lesser role in predicting well-being while controlling for romantic involvement and other close relationships (Bertera 2005; Walen and Lachman 2000; Whisman et al. 2000). This is not to deny the role of friends in one’s life. Friends add joy to the lives of people but their contribution might be less transparent and obvious when achieving an important task of emerging adulthood which is establishing a romantic relationship. Secondly, from a different perspective, singles might still benefit from their close relationships, especially from their friendships. Decades of research suggests that establishing and maintaining romantic relationships might be the key for happiness (DePaulo and Morris 2005, Seligman 2002; Myers 1992). Even though this might be the case, findings suggest that among those without a romantic partner, relationship experiences with best friends and mothers play an important role in the happiness of emerging adults. Finally, even though friends or romantic partners might consume the top positions in one’s hierarchy of close relationships, it seems that relationships with family members, especially with the mother, play an important role in happiness among emerging adults. The results showed that mother–child relationship quality emerged as a significant predictor of happiness when compared to father–child relationship quality in the two groups compared (with/out a romantic partner). This suggests that even though emerging adults experience changes in their relationships with their parents (Furman and Buhrmester 1992; Wel et al. 2000), relationship experiences with mothers continue to play an important role in happiness. These findings could be considered to provide support for the arguments that relationships with mothers as compared to fathers have a stronger influence on well-being (e.g., Field et al. 1995).

It is imperative to note that these findings do not suggest that the relationship with the father is not important for the happiness of emerging adults. As reported in Tables 2 and 3, the quality of father relationships was related to happiness among emerging adults with and without a romantic partner at the bivariate level. Findings only suggest that father relationships might not emerge as a significant predictor of happiness when simultaneously considered with the other close relationships of the emerging adults.

There is one point that needs to be highlighted considering the findings obtained in the present study (we thank the anonymous reviewer for this comment). According to some scholars, there is one common but unrecognized stereotype in the American culture suggesting that being in a romantic relationship (dating or married) is superior to being single (DePaulo and Morris 2005). This stereotype holds that singles who are not involved in a romantic relationship are self-centered, lack positive social relationships and experience lower levels of well-being. However, DePaulo (2006) and her colleague (DePaulo and Morris 2006), debunked these myths, after a critical analyzes of the empirical literature, by showing that singles were equally likely to pursue their goals and be successful in their lives, and be healthy and happy, compared to their romantically involved peers. Supporting these findings, in the present study those with and without a romantic partner did not differ from each other on happiness (t (309) = .333, NS). More importantly, single emerging adults were more likely to benefit from the quality of their best friendships compared to those involved in a romantic relationship. It seems that being single might present certain benefits to the individual. Accordingly, it seems reasonable to suggest that being in a romantic relationship is not superior to being single, at least for the participants of this study. Future research would greatly benefit by comparing the lives of single and romantically involved emerging adults in different domains.

The findings of the present study might also have some implications for clinical practice. First of all, counselors in college settings might benefit by inquiring about the close relationships experiences of emerging adults. Secondly, learning about the relationships status (single vs. dating) of the emerging adult might be important since the findings suggest that close relationships might play a different role in well-being depending on whether one is single or in a romantic relationship. In the case of single emerging adults, helping them improving their existing friendships might be a good strategy. Indeed, establishing and maintaining positive friendship experiences with friends are found to be one of the strongest predictor of adjustment in college (Devlin 1996; Paul and Brier 2001; Paul and Kelleher 1995; Dixon Rayle and Chung 2007–2008). In the case of romantically involved emerging adults, the priority might be directed to improve the overall relationship experiences with the romantic partners (e.g., teaching skills to handle relationship conflict effectively). This is because, as the findings suggests, emerging adults experience higher levels of happiness when their romantic experiences are higher in overall quality. At the same time, managing conflicts effectively and/or experiencing lower levels of conflict with the romantic partner might help the emerging adult to benefit from other close relationships as well (e.g., best friends).

One essential point we would like to highlight is whether or not positive and negative affect and life satisfaction constitute a higher order factor of happiness. Theory suggests that these variables make up the construct of subjective well-being (Diener 1984; Diener 1994). Although the three components are different constructs assessing the cognitive and affective dimensions of subjective well-being, they are substantially correlated and research suggests a higher order factor (Diener et al. 1999; Diener 1994), which is referred to as happiness. Specifically, several published studies, in this journal and others, report that factor analyzes (after recoding negative mood) reveals a single factor accounting for as high as 56% of the variance (Bettencourt and Sheldon 2001; Demir and Weitekamp 2007; Elliot et al. 1997; Kasser and Sheldon 2002; Sheldon and Elliot 1999; Sheldon and Hoon 2007; Sheldon and Kasser 1995). Accordingly, studies standardize the measures and create an aggregate happiness score by summing positive affect and life-satisfaction, then subtracting negative mood (e.g., Sheldon et al. 2005). This could be a valid way of creating an aggregate happiness score. However, future research could examine whether the findings are replicated when the components of happiness are examined separately.

5.4 Limitations

The present investigation was not without limitations. First of all, the results reflect the responses of emerging adults attending a large Midwestern university. As such, the findings cannot be generalized to emerging adults not in college or other age groups. Second limitation involves not assessing the status of romantic relationship (dating, cohabiting, married). Would the findings obtained in the present study be similar among emerging adults in different type of romantic relationships? It is the task of future research to investigate this among emerging adults in different levels of romantic involvement. Third limitation involves not assessing the sexual orientation of the individuals when assessing romantic relationships. This raises the possibility that straight and lesbian/gay/bisexual and transgendered (LGBT) individuals might have participated in the study. Considering the empirical research showing that the close relationship experiences of LGBT individuals might be qualitatively different compared to straight individuals (Blair and Holmberg 2008; Kurdek 2004; Muraco 2003; Nardi 1999; Nardi and Sherrod 1994), it would be interesting to examine if the findings reported in this study would be observed among emerging adults with different sexual orientations. Finally, the present study focused on four close relationships. Although the relationships investigated might capture the most significant relationships in one’s life at least among those in college and who might not have a children of their own, individuals have close relationships with their siblings (Cicirelli 1995) and relatives (e.g., uncles) (Hinde 1997) as well, and these relationships are also related to well-being (e.g., Sherman et al. 2006). Future research might greatly benefit by considering other close relationships of the individual in predicting happiness among emerging adults with and without a romantic partner.

6 Conclusion

The present study investigated the role of multiple close relationships in the happiness of single and romantically involved emerging adults. Findings revealed that the quality of mother and best friend relationships were the only predictors of happiness among emerging adults without a romantic partner, whereas the quality of mother and romantic relationships emerged as the important predictors of happiness among those involved in a romantic relationship. These findings suggest that emerging adults with a romantic partner might benefit less or not at all from the quality of their friendships when close relationships are considered simultaneously.

Acknowledgments

The author is appreciative of the comments of the anonymous reviewers. The author is also grateful to Adilşah Demir, Metin Őzdemir and Őzgür Parlak for their helpful comments while preparing the manuscript.

Copyright information

© Springer Science+Business Media B.V. 2009