Happiness n Friendship

MEL_IKSAH DEM_IR and LESLEY A. WEITEKAMP

I AM SO HAPPY CAUSE TODAY I FOUND MY FRIEND:

FRIENDSHIP AND PERSONALITY AS PREDICTORS

OF HAPPINESS

(Received 12 April 2006; Accepted 16 August 2006)

ABSTRACT. Relying on the theoretical model of [Lyubomirsky et al. 2005,Review of General Psychology, 9, pp. 111–131], the present study investigatedthe relationship between personality, number of friends, best friendship qualityand happiness among 423 young adults (n = 300 women). The main interestwas to examine whether friendship contributed to happiness while controllingfor personality. Friendship variables accounted for 58% of the variance inhappiness. Results revealed that friendship quality predicted happiness aboveand beyond the influence of personality and number of friends, but friendshipconflict was not a significant predictor. Additional analyses revealed that thecompanionship and self-validation features of friendship quality were predic-tive of happiness while controlling for gender and personality. The findingswere discussed in the light of theory and empirical research and suggestionswere made for future research.

KEY WORDS: companionship, friendship, happiness, personality, self-validation

INTRODUCTION

What makes people happy? This has become a popular ques-tion in psychology and a burgeoning body of research docu-mented factors that predict happiness (for a review see Dieneret al., 1999). In the present study, we focused on two of thevariables that have been consistently associated with happiness:friendship and personality. We had two aims; first, we investi-gated the predictive ability of best friendship quality andconflict for happiness above and beyond the influence of per-sonality. Then, if friendship quality proved to be a significantpredictor, we would examine which dimension(s) of friendship(e.g., intimacy) were the most important predictor of happiness.

Journal of Happiness Studies (2007) 8:181–211 � Springer 2006DOI 10.1007/s10902-006-9012-7

Happiness: Definition and Factors Influencing Happiness

It is essential to describe happiness before we review the rela-tionship between happiness and the study variables. Happiness,or subjective well-being, as it is interchangeably used in the lit-erature, refers to the cognitive and affective evaluations of one�sown life (Diener, 1984, 1994). Happiness is defined in terms ofglobal life satisfaction, presence of positive affect and absenceof negative affect. Although these three components are differ-ent constructs and might require further research to understandeach of them separately (Diener et al., 1999), the three compo-nents are substantially correlated and research suggests a higherorder factor (Diener, 1994; Sheldon and Lyubomirsky, in press),which is referred to as happiness. We measured happiness as itwas commonly assessed in the literature.

Research suggests that there are several factors that predicthappiness. Recently, Lyubomirsky et al. (2005) proposed amodel that attempts to account for factors influencing the cur-rent happiness level of individuals. We relied on this model as aframework to set out variables in a meaningful way in predict-ing happiness. Accordingly, within this model there are threemain factors that influence happiness, these are: the happinessset point, circumstances and intentional activities.

The first factor that influences happiness, set point, refers tothe ‘‘the central or expected value within the person�s set range(Lyubomirsky et al., 2005, p. 116).’’ Set point for happiness isassumed to be heritable, fixed, and stable over time. Indeedthere is research evidence supporting these assumptions(McCrae and Costa, 1990; Tellegen et al., 1988). According tothe model, set point accounts for 50% of the variance in happi-ness. Among the variables posited to be under this dimension,we focused on personality and relied on the Big Five frameworkto assess personality.

The second factor that influences happiness, circumstances,refers to the ‘‘incidental but relatively stable facts of anindividual�s life (Lyubomirsky et al., 2005, p. 117).’’ Theseinclude demographic (e.g., gender), geographic (e.g., where oneresides) and contextual (e.g., culture) variables. There isresearch showing that gender, income of the individual, marital

MEL_IKSAH DEM_IR AND LESLEY A. WEITEKAMP182

status, and culture makes a difference in happiness. However,as the literature reviews have documented, circumstantial/contextual factors account for about 10% of the variance inhappiness (Diener et al., 1999).

The third factor that influences happiness, intentional activi-ties, is a broad category that involves the voluntary and effort-ful things people do in their everyday lives. Lyubomirsky et al.(2005) differentiates between three types of intentional activities:cognitive, behavioral, and volitional. Cognitively, one could becontemplating about good things going on in his/her life.Behaviorally, one could provide instrumental support to his/herfriend. Volitionally, one could develop personal goals and striveto achieve them. According to Lyubomirsky et al. (2005), thecommon theme across the three types of activities is that all relyon the voluntary effort of the individual. The research ofLyubomirsky and her colleagues and research they reviewedsuggest that intentional activities account for about 40% of thevariance in happiness (Lyubomirsky et al., 2005; Sheldon andLyubomirsky, 2006b). Supporting the proposition of the model,Csikszentmihalyi and Hunter (2003) showed that what an indi-vidual does and with whom has the potential to change the setpoint of happiness.

Relationships are not clearly specified by the model as beingunder the rubric of intentional activities, however, we believerelationships could be considered under this dimension and fo-cused on best friendship in the present study. We deem thatactivities are not practiced just by the individual alone butcould include others as well. Our belief that friendship could beconsidered under this dimension stems from the well-acceptedargument that friendships are voluntary relationships and in-volves a variety of activities. Indeed, the activities one engagesin with friends (e.g., talking on the phone) are considered underthe rubric of social activities (Argyle and Lu, 1990). For in-stance, as it will be considered below, doing things togetherwith a friend is an important dimension of friendship. Our con-sideration for including friendship under this dimension is notonly behavioral, but cognitive as well. Cognitively, one mightthink about the positive experiences with a friend, which could

FRIENDSHIP AND HAPPINESS 183

in turn make the person feel good about him/herself. Overallthe inclusion of friendship as a factor in the intentional activi-ties domain seems justified.

It is important in this point to consider recent theoreticaladvances in the field. The first factor of the model ofLyubomirsky et al. (2005) considers set point as an importantfactor for the baseline happiness. A recent paper by Dieneret al. (2006) reviewed considerable evidence showing that it ispossible to observe changes in the set point for happiness. Theactivities people engage in (Sheldon and Lyubomirsky, 2004),happiness interventions (Seligman et al., 2005) and changes inmarital status (see Diener et al., 2006) can all modify the setpoint for happiness. Testing the predictive ability of friendshipabove and beyond the influence of personality offered an excel-lent opportunity to examine whether friendship has the poten-tial to modify the set point for happiness.

In summary, our focus was on personality from the set pointdimension and best friendship from the intentional activitiesdimension. Each one of the variables and their relationship tohappiness is reviewed below.

Personality and Happiness

Research has shown that all dimensions of the Big Fiveframework are associated with happiness to differing degrees(Compton, 1998; DeNeve and Cooper, 1998; Furnham andCheng, 1997; McCrae and Costa, 1991; Morrison, 1997; Ruiz,2005). Among the big five, extroversion and neuroticism wereconsistent predictors, extroversion being positively and neuroti-cism being negatively correlated to happiness (Argyle, 2001;Cheng and Furnham; 2001). It was also found that conscien-tiousness, agreeableness and openness to experience were slight-ly positively correlated with happiness (for a review see Dieneret al., 1999). Together, these personality dimensions account forbetween 20 and 50% of the variance in happiness. Since previ-ous research was not consistent in finding a link between con-scientiousness, agreeableness and openness to experience, we didnot have any specific hypotheses for the relationship betweenthese dimensions and happiness.


Friendship

Relationships, and especially friendships, play an integral partin our everyday life. Everybody talks about friendship andmany have their own way of describing it, however, from anempirical standpoint, what is friendship? Although scholarsagree that it is not easy to define friendship, empirical literatureobserved several attempts to conceptualize friendship (for a re-view see Fehr, 1996). The definition provided by Hays (1988)captures all of the previous conceptualizations and definesfriendship as a ‘‘voluntary interdependence between two personsover time, which is intended to facilitate socio-emotional goalsof the participants, and may involve varying types and degreesof companionship, intimacy, affection and mutual assistance(p. 395).’’ In other words, friendship is a qualitative relationshipbetween two individuals (Hinde, 1997).

Implicit in the definition above is the variety of features, orso called provisions, friendship involves. In the presentstudy, we used the McGill Friendship Questionnaire-Friend�sFunctions (MFQ-FF) to assess overall relationship quality(Mendelson and Aboud, 1999). MFQ-FF is designed for usewith adults and it assesses six theoretically identified features offriendship. These features are companionship, help, intimacy,reliable alliance, self-validation, and emotional security. Com-panionship is considered to be an important dimension offriendship across all ages and involves doing things together.Help refers to the guidance, assistance and aid one receives.Intimacy involves experiencing a context where one could open-ly and honestly disclose personal and private information. Reli-able alliance refers to trust and loyalty; in essence knowing thatyour friend will be there no matter what. Self-validation refersto perceiving the other as supporting and encouraging of one�simage, success, and self. Finally, emotional security refers to theother�s providing comfort and giving reassurance in new andthreatening situations. The common practice in the literature isto sum the means of different features and the overall score iscalled friendship quality.

Relationship scholars agree that friendship is a mixed bless-ing that can at times involve conflict (Hinde, 1997). Research


done with children and adolescents suggest that individuals doindeed engage in conflict with their friends (Hartup andStevens, 1997), and conflict has been argued to be an inevitabledimension of relationships (Hinde, 1997). Considering the theo-retical arguments and empirical research, we assessed bestfriendship conflict as well.

Individuals are likely to have several friends but it isimportant to differentiate degree of closeness. People makeclear distinctions between best, close and casual friendships(Fehr, 1996; Hays, 1988). Indeed, there is empirical evidenceshowing that individuals have higher quality relationshipswith their best friends as compared to less close friends(Davis and Todd, 1985; Wright, 1985). In other words, themore intimate the friendship the more clearly it manifeststhe various attributes of friendship (Fehr, 1996, p. 19). Sincethe empirical literature shows that the quality of bestfriendships is different from other friendships, we focusedexclusively on best friendships.

One final point to consider is the number of friends anindividual has. Research in different fields simply asks indi-viduals how many friends they have and uses this as an indi-cator of friendship. Number of friends indeed is included insome models of friendship (Bukowski and Hoza, 1989), how-ever, considering the theoretical arguments that friendship isa qualitative relationship (Hinde, 1997), knowing how manyfriends one has does not provide any insight about the qual-ity of the friendships. We believe that at best, number offriends reflects the sociability and likeability of the individual.Since number of friends has been considered in certain theo-retical models and was a focus of interest in some studies onhappiness (Burt, 1987; Myers, 2000), we investigated and con-trolled for the relationship between number of friends andhappiness.

Friendship and Personality

Research using the Big Five personality framework showed thatall dimensions of personality (extroversion, agreeableness, open-ness to experience, and neuroticism) except conscientiousness


were related to friendship in some way. The most consistentrelationships found across studies have been between extrover-sion and agreeableness and friendship. A number of studiesdocumented that extroverts were more likely to develop a net-work of friends in their first year in college (Asendorpf andWilpers, 1998) and had more friends than introverts (Hills andArgyle, 2001). Other studies also reported that extroversion wasrelated to the closeness felt toward friends (Berry et al., 2000;Jensen-Cample et al., 2002; Wilkinson and Watford, 2001).Agreeableness as a personality dimension was also investigatedin relation to friendships. Several studies reported that thisdimension of personality was related to better functioningfriendships (Asendorpf and Wilpers, 1998; Berry et al., 2000;Jensen-Cample et al., 2002). Those individuals high in agree-ableness had higher quality and lower conflict friendships. Asfor neuroticism and openness to experience, Wilkinson andWatford (2001) found that neuroticism and feelings of closenessto friends were moderately negatively correlated, and Berryet al. (2000) reported that these two dimensions were related tothe number of conflicts reported between friends in a week.

Friendship and Happiness

Research investigating the relationship between friendship andhappiness has considered friendship in different ways. For in-stance, Baldassare et al. (1984) reported an association betweensupport received from a friend, companionship, and happinessand found that perceived companionship was the best predictorof happiness. Gladow and Ray (1986) also reported thatsupport from friends was related to happiness. Diener andSeligman (2002) and Lyubomirsky et al. (2006) showed thatcloseness and satisfaction with a friend was associated with hap-piness. Important to note is that closeness and satisfaction weremeasured with one item. We believe that assessing a qualitativerelationship with one item does not provide enough informationabout the quality of friendship. Although the research docu-menting an association between friendship and happiness isimportant, it is limited by not investigating the relationship ofdifferent dimensions of friendship (quality and conflict), as


conceived in the literature, to happiness. As for friendship con-flict and happiness, we identified two studies that documented anegative association between the two among late adolescents(Mathur, 2004). Research investigating the link between conflictand other adjustment indices (e.g., depression) also reported anegative association (Demir and Urberg, 2004). Thus, a consid-erable amount of research has established that friendship qual-ity and conflict is related to happiness.

In the present study, we first tested whether friendship qual-ity and conflict were related to happiness. Second, we examinedthe predictive ability of overall friendship quality and conflicton happiness while controlling for personality (see below). If wefound evidence for the predictive ability of friendship quality,then, we were interested in examining which feature(s) of friend-ship predicted happiness above and beyond the impact of gen-der and personality. In other words, the analyses pertaining tothe second question were conditional on finding an incrementalpredictive ability of overall friendship quality on happiness.

Controlling for personality while testing the role of friend-ship in happiness is an important aspect of this study, and de-serves attention. We believe that one major limitation of theavailable literature that documents an association betweenfriendships and happiness is that they do not control for per-sonality, which explains a significant amount of the variance inhappiness. It could be that the relationship between social expe-riences and happiness might disappear once personality is takeninto account. Are there any theories that might offer a predic-tion? Other than major meta-theories positing that friendshipsare important for happiness (Baumeister and Leary, 1995), wedo not know of any specific hypothesis regarding the role ofrelationships in happiness above and beyond the influence ofpersonality. Nevertheless, we identified one longitudinal studythat reported the predictive ability of social support in predict-ing happiness above and beyond the influence of baseline happi-ness and personality (extroversion and neuroticism) (Lu, 1999).Although important, this study assessed only social support (thesource of support was not clear) and was done with adults ofdifferent ages in Taiwan. It remains to be tested whether the


findings of Lu (1999) could be obtained for friendship qualityor among American college students.

Another way to explore the role of friendship in happinesswhile controlling for personality would be to pay attention tothe studies on extroversion. A considerable amount of researchinvestigated the relationship between extroversion and happi-ness (see Diener and Lucas, 1999). The research of Lucas andcolleagues suggests that (Lucas and Diener, 2001; Lucas et al.,2000) extroversion and an extrovert�s greater sensitivity to re-wards might be the main factor predicting happiness. Followingthis, one might argue that the greater number of friends extro-verts have and their closer relationships with best friends (Hillsand Argyle, 2001) might be attributed to their sensitivity to re-wards and these would not emerge as predictors of happiness.The study reported by Lu (1999) suggests that this could not bethe case. In the present study controlling for personality whentesting the predictive ability of friendship quality offered anexcellent opportunity to investigate these claims.

It must be noted that the predictive ability of friendships wassubjected to a strict test in the present study. Philosophical andtheoretical arguments that span over nearly two millennia attri-bute friendship with having an important role in well-being. Afailure to find incremental predictive ability for friendship inhappiness while controlling for personality would not necessar-ily mean friendship is not important for well-being, rather, itwould mean that friendship is not important for happinesswhile controlling for set-point variables. Friendship might beimportant for other well-being indices (e.g., depression) and assuch it should be the task of future research to investigate thispossibility. Finding that friendship still predicts happiness whilecontrolling for personality would add confidence to the philo-sophical and theoretical arguments and confirm and extend theargument made earlier that friends add something extra to lifeeven when you have your set point for happiness.

One final note is on gender. Considering the research show-ing that gender was weakly related to happiness, it was not ourfocus in the present study. We nevertheless controlled for gen-der in our analyses since a considerable amount of research


documented gender differences in friendships (Fehr, 1996), withwomen having friendships high in quality and low in conflict ascompared to men.

Considering the available research reviewed above, we devel-oped the following hypotheses:

(a) Friendship quality would be positively and conflict nega-tively correlated with happiness.

(b) Extroversion (positively) and neuroticism (negatively)would be predictors of happiness.

(c) All personality dimensions with the exception of conscien-tiousness and openness would be related to friendshipquality.

(d) All personality dimensions with the exception of conscien-tiousness would be related to friendship conflict.

(e) Friendship quality and conflict would be predictive ofhappiness above and beyond the influence of gender andpersonality.

METHOD

Participants

Our sample consisted of 423 (300 women, 123 men) studentsattending a Midwestern university. The age of the sampleranged from 18 to 44 years of age (M = 22.53, SD = 4.69).The ethnic distribution of the sample was as follows: 48%Caucasian (n = 203), 32.6% African-American (n = 138), 8%Asian (n = 34), and other 11.4% (n = 37).

Procedure

A psychology student pool was used to recruit participants.Announcements were made in classrooms and flyers were pos-ted in the psychology department. Those who wanted to partici-pate in the study either took the survey with them to completeon their own time or completed the questionnaire packet in ourlab. The packet included a consent form, a basic demographicinformation sheet and a battery of questionnaires. To ensureprivacy, participants were given envelopes to enclose the


completed surveys. Those taking the surveys with them placedthe envelopes in a designated location or turned them in di-rectly to the researcher. Completion of the survey lastedapproximately 30 min and participants earned extra credit fortheir psychology classes.

Measures

Best Friendship QualityMcGill Friendship Questionnaire-Friend�s Functions (MFQ-FF,Mendelson and Aboud, 1999) was used to assess best friendshipquality. MFQ-FF consists of 30 items, five for each of the sixfunctions assessed: stimulating companionship, help, intimacy,reliable alliance, emotional security, and self-validation. Sampleitems include ‘‘my best friend is fun to sit and talk with’’ and‘‘my best friend is someone I can tell private things.’’ Items wererated on a 9-point scale (0--8) on which five of the points arelabeled (0 = never, 2 = rarely, 4 = once in a while, 6 = fairlyoften, and 8 = always). The mean of 30 items was taken to forman overall Best Friendship Quality composite score. The reliabil-ity of the scale in the present study was high (a = 0.95).

Best Friendship ConflictFor the purposes of this study, five items were written to assessBest Friendship Conflict. Items were taken from existing scales(Bukowski et al., 1994; Parker and Asher, 1993). The items usedin the present study were: ‘‘my best friend and I have manyquarrels;’’ ‘‘my best friend and I argue a lot;’’ and ‘‘my bestfriends can bug me or annoy me even though I ask him not todo;’’ ‘‘my best friend makes me upset or mad’’ and ‘‘my bestfriends and I disagree a lot.’’ Items were rated on a 9-pointscale (0–8) on which five of the points are labeled (0 = never,2 = rarely, 4 = once in a while, 6 = fairly often, and 8 =always). The five items were summed to create the Best Friend-ship Conflict composite score. The reliability of the conflictscale in the present study was high (a = 0.88).

In our study, 11.1% (n = 47) of the participants reportednot having a best friend and did not complete the questionnairefor best friendship quality and conflict. Therefore, our final


sample consisted of 376 (268 female) participants who reportedhaving a best friend.

PersonalityThe Big Five Inventory (BFI) (John and Srivastava, 1999) wasused to assess personality. The BFI consists of 44 items andassesses five personality dimensions (extroversion, agreeableness,conscientiousness, neuroticism, and openness). Participants areasked to rate their agreement with each BFI item on a 5-pointscale ranging from 1 (disagree strongly) to 5 (agree strongly).Sample items include ‘‘I see myself as someone who is talka-tive’’ and ‘‘I see myself as someone who is get nervous easily.’’The reliabilities of the scales in the present study were as fol-lows: a = 0.82 for extroversion; a = 0.75 for agreeableness;a = 0.77 for conscientiousness; a = 0.76 for neuroticism; anda = 0.74 for openness.

HappinessFollowing the literature, we assessed happiness with the Satis-faction with Life Scale (SWLS) (Diener et al., 1985a) and thePositive and Negative Affect Schedule (PANAS) (Watson et al.,1988). Prior studies also relied on these instruments to assesshappiness (Sheldon and Kasser, 2001; Sheldon et al., 2005).

The SWLS assesses the global cognitive evaluations of one�slife. The scale consists of five items and respondents are askedto rate their agreement with the items on a 7-point scale rang-ing from (1) strongly disagree to (7) strongly agree. The meanscores for the five items will be used to obtain a satisfactionwith life composite score. The reliability of the scales for thepresent study was high (a = 0.86).

The PANAS was used to assess general positive and negativeaffect. The PANAS consists of 10 mood states for positiveaffect (PA) (e.g. attentive) and 10 for negative affect (NA)(e.g. hostile). Respondents are asked to rate the extent to whichthey feel each mood in general on a 1 (very slightly or not all)to 5 (extremely) scale. The mean scores for the 10 positive andnegative affect scores were used to create a composite positiveand negative affect scores. Reliabilities for the scales were satis-factory (a = 0.82 for PA; a = 0.84 for NA).


In order to simplify the presentation of happiness, we createdan aggregate happiness score by standardizing the satisfactionwith life, positive and negative affect scores and subtracting thestandardized negative affect scores from the sum of standardizedsatisfaction with life and positive affect scores. This procedure re-sulted in scores that ranged from )7.12 to 4.80. Creation ofaggregate scores of happiness have been reported by others aswell (Kasser and Sheldon, 2002; Sheldon and Kasser, 2001). Wewanted to compare the range of happiness score to other studies;however, not every study reports this information. Nevertheless,the range found in the present study was similar to what wefound in our other studies (Demir et al., 2006b). Also, the meansand standard deviations reported for measures used were similarto those found in other research (Kasser and Sheldon, 2002).

The practice of allowing some participants to complete thesurveys at home might introduce some confounds to the study.In the present study, 240 participants completed the surveys atour lab and 136 participants took the survey with them andreturned it later. In order to test if this was the case we firstcompared the mean scores of each variable across the twogroups. The two groups significantly differed from each other onagreeableness (t(374) = 2.705, p<0. 05) and openness to expe-rience (t(374) = 3.571, p<0.01). Those taking the survey at ourlab scored higher on agreeableness (mean = 4.07 vs. 3.89) andopenness to experience (mean = 3.57 vs. 3.29) than those takingthe survey with them and returning later (mean = 3.89). As asecond step, we computed the regression analyses reported be-low for the two groups and the results were almost the same(slightly differing b values) with the exception of friendship qual-ity marginally (p = 0.8) predicting happiness in the secondgroup. This could be attributed to the smaller number of casesin this group. To sum up, the way the data collected did notseem be a confounding variable for our results.

RESULTS

Descriptive statistics and correlations for the study variables arereported in Table I. As seen from the table, gender was not


TABLE

IDescriptivestatisticsandcorrelationmatrix

MSD

12

34

56

78

910

11

12

13

1.Gender

1.71–

–2.Extroversion

3.440.75

0.05

–3.Agreeableness

3.990.63

0.09

0.25*

–4.Conscientiousness2.730.75

0.21*

0.15*

0.31*–

5.Neuroticism

3.760.62

0.15*

)0.38*

)0.36*

)0.27*–

6.Openness

3.590.60

0.07

0.15*

)0.05

0.10*

)0.07

–7.Friendship

quality

6.760.92

0.24*

0.21*

0.34*

0.16*

)0.09

0.01

–

8.Friendship

conflict

2.511.62

)0.11**

)0.11**

)0.35*

)0.18*

0.23*

0.02

)0.27*

–

9.Number

of

friends

4.4

2.6

)0.13**

0.13**

0.01

0.02

)0.14**0.05

0.10**

)0.06

–

10.Positiveaffect

3.820.58

0.06

0.45*

0.37*

0.35*

)0.46*

0.16*

0.25*

)0.17*

0.16*–

11.Negativeaffect

2.040.65

)0.05

)0.32*

)0.40*

)0.30*

0.71*

0.01

)0.16*

0.30*

)0.10

)0.34*–

12.Lifesatisfaction

4.651.28

0.10

0.31*

0.25*

0.19*

)0.34*

0.01

0.18*

)0.17*

0.15*

0.44*

)0.34*–

13.Happiness

0.072.30

)0.10

0.48*

0.44*

0.37*

)0.66*

0.06

0.26*

)0.28*

0.19*

0.77*

)0.73*0.78*–

*p<

0.05,**p<

0.01.


correlated with happiness but was related to friendship vari-ables. Considering this, we controlled for gender in the analysesreported below.

Supporting our hypothesis (a), friendship quality and conflictwere related to happiness in the expected directions. Zero-ordercorrelations between personality and happiness revealed that allpersonality variables with the exception of openness to experi-ence were significantly correlated with happiness, with neuroti-cism having the strongest relationship. These findings supportedour hypothesis (b) for extroversion and neuroticism, and alsoshowed that agreeableness and conscientiousness were related tohappiness in the present sample.

Our hypothesis for friendship quality and personality (c) waspartially supported in that only extroversion, agreeableness andconscientiousness were significantly correlated with friendshipquality. Our hypothesis for friendship conflict (d) was also par-tially supported. All personality dimensions with the exceptionof openness to experience were significantly correlated withfriendship conflict. Across both friendship variables, agreeable-ness had the strongest relationship.

Does Friendship Predict Happiness Above and Beyond

the Influence of Personality?

We relied on multiple hierarchical regressions in testing the mainquestion of the present study. In predicting happiness, genderwas entered in the first step as a control variable, the second stepincluded personality variables, number of friends was entered inthe third step and the fourth step included the friendship vari-ables. We also computed interaction terms for gender andfriendship variables and included them in the fourth step.

Results showed that gender, entered in the first step, wasmarginally significant and accounted for 1% of the variancein happiness (F(1, 374) = 3.483, p<0.06). b values for thevariables are reported in Table II. The second step involvingpersonality variables was also significant (F(5, 369) = 92.724,p<0.01) and accounted for an additional 55% of the variancein happiness. As it can be seen from Table II, openness to expe-rience was not significant and of the four significant predictors,


neuroticism had the highest b value. The third step involvingthe number of friend was not significant (F(1, 368) = 2.147,ns). The fourth step involving friendship variables was also sig-nificant and explained an additional 2% of the variance in hap-piness (F(1, 366) = 6.341, p<0.01). As seen in Table II, onlyfriendship quality predicted happiness above and beyond thepersonality variables. This finding partially supported ourhypothesis (e). The last step including interactions was not sig-nificant and therefore not reported.

As specified in the text, our interest was in the predictiveability of friendship while controlling for personality. In orderto provide information to the reader, we computed an addi-tional analysis in which friendship quality and conflict wereregressed on happiness while controlling for gender. Resultsshowed that both friendship variables predicted happiness(F(3, 372) = 20.889, p<0.01) and accounted for 15% of thevariance.

TABLE IIHierarchical multiple regression predicting happiness (n = 376)

b

Step 1Gender (1 = male, 2 = female) )0.13

R2 = 0.01**Step 2Extroversion 0.23*Agreeableness 0.13*Conscientiousness 0.17*Neuroticism )0.44*Openness )0.01

DR2 = 0.55*Step 3Number of friends 0.06

DR2 = 0.55Step 4Best friendship quality 0.12*Best friendship conflict )0.06

DR2 = 0.02*

Note: b weights are for the final model.*p<0.06, **p<0.01.


Which Friendship Quality Dimension(s) Predict Happiness?

Since friendship quality appeared to be a significant predictor ofhappiness, we computed another multiple regression to identifywhich dimension(s) of friendship was predictive of happinesswhile controlling for gender and personality. Gender was en-tered in the first step as a control variable, followed by person-ality variables in the second step, and companionship, help,intimacy, reliable alliance, self-validation and emotional securitywere entered in the third step.

The results for gender and personality were similar to thefirst regression with the exception of neuroticism having a bvalue of )0.45 instead of )0.44. The third step includingthe friendship quality dimensions was significant (F(12, 363) =42.886, p<0.01) and explained an additional 3% of variance inhappiness. As it can be seen from Table III, of the six dimen-sions of friendship only companionship and self-validation weresignificant predictors of happiness controlling for gender andpersonality.

DISCUSSION

The present study investigated the relationship betweenpersonality, friendship and happiness. At the bivariate level,

TABLE IIIHierarchical multiple regression of friendship qualities on happiness

(n = 376)

ba

Companionship 0.10*Help 0.03Intimacy 0.05Reliable alliance 0.04Self-validation 0.12*Emotional security 0.01

Note: b weights are for the final model.*p<0.05.aControlling for gender and personality.


findings fully or partially supported our hypotheses. All person-ality variables but openness to experience were significantlycorrelated with happiness. Supporting previous research, neu-roticism was found to have the strongest relationship with hap-piness (Furnham and Cheng, 1997; Ruiz, 2005). Our findingsfor friendship quality and conflict were consistent with the liter-ature (Asendorpf and Wilpers, 1998; Jensen-Cample et al., 2002;Wilkinson and Watford, 2001). When analyzing our results,three points are important. First, consistent with the literature(Berry et al., 2000), neuroticism was related to friendship con-flict but not quality. It seems that high scorers on neuroticismdo not necessarily have lower relationship quality, but are likelyto experience high levels of conflict. Second, agreeableness hadthe strongest relationship with friendship quality and conflict.Consistent with the literature, those scoring high on agreeable-ness had an ideal friendship in that quality was high and con-flict was low (Berry et al., 2000; Jensen-Cample et al., 2002).This finding also lends support to the theoretical arguments thatagreeableness facilitates more positive experiences in social situ-ations (McCrae and Costa, 1991). Finally, supporting previousresearch (Wilkinson and Watford, 2001), openness to experiencewas not related to friendship variables. Finally, as for the rela-tionship of friendship and happiness, we found that both friend-ship quality and conflict were related to happiness. Thesefindings extend prior research in that theoretically developedmeasures of friendship were related to happiness.

Predicting Happiness

In the present study, we investigated the predictive of abilityof number of friends, best friendship quality and conflict inhappiness while controlling for personality. The Big Fivedimensions accounted for 55% of the variance in happiness.This is consistent with the literature and confirms the well-established arguments once more that personality is a strongpredictor of happiness (Diener et al., 1999; Lyubomirsky et al.,2005). The results also showed that even though number offriends was related to happiness at the bivariate level, it wasnot predictive of happiness above and beyond the influence


personality. This suggests that even though one might have sev-eral friendships it is the quality of friendship but not quantitythat makes a difference in the happiness of the individual. Thisfinding might not be surprising when one considers the fact thatfriendship is a qualitative relationship (Hinde, 1997) and thenumber of friends one claims to have does not inform us aboutthe quality of those friendships. Finally, controlling for person-ality and number of friends, it was only friendship quality thatpredicted happiness and accounted for 2% of the variance. Webelieve this finding is important and deserves attention.

Friendship is an important relationship and is cherished notonly by theorists but also by the general population. From phi-losophers to social scientists, friendship was attributed severalimportant functions and one of them was its role in happiness.We subjected friendship to a strict test in that we controlled itspredictive ability while controlling for personality. This was thestrength of the present study in that previous research docu-menting the link between social experiences and happiness didnot control for personality. We found support for these argu-ments such that friendship quality, but not conflict, made anadditional contribution to happiness even when controlling forpersonality. This suggests that even though one could havecertain predispositions to be happy, having a friendship that ishigh in quality still adds something extra to our lives and hasthe potential to increase one�s happiness level. In other words,the relationship between friendship and happiness cannot beaccounted for by personality. These findings are consistent withresearch conducted in other cultures (Lu, 1999) and recent theo-retical advances that point to the modifiability of the set pointfor happiness (Diener et al., 2006). We hope this findingwould inspire relationship scholars to develop specific hypothe-ses about the role friendships play in the well-being of theindividuals.

In reviewing the work on extroversion (Lucas and Diener,2001), we highlighted the possibility that the greater number offriends extroverts have and their closer relationships with bestfriends (Hills and Argyle, 2001) might be attributed to their sen-sitivity to rewards but this would not appear a predictor of


happiness. Our results revealed that this was not the case atleast in the present study. This suggests that experiences withfriends, but not number of friends have the potential to contrib-ute to happiness above and beyond extroversion and sensitivityto rewards. Future studies are needed to replicate this finding inorder to establish confidence in the results reported.

Even though we found that friendship quality contributed tothe happiness of the individual above and beyond the influenceof one�s personality, it explained only an additional 2% of thevariance in happiness. Is this an important finding or could beconsidered as negligible? We do believe that this is a significantfinding for two reasons. First of all, when considered alonefriendship quality and conflict together explain 15% of the vari-ance in happiness. There are several variables that predict hap-piness and we believe the amount of variance accounted for byfriendship is substantial if not impressive. Second of all, friend-ship quality was still a predictor of happiness while controllingfor personality, a major predictor of happiness. We believe thatif friendship were not important for the happiness of the indi-vidual, then, it would not contribute to happiness above andbeyond the influence of set point variables. Findings suggestthat friendship is indeed important regardless of one�s personal-ity. The way one interprets the 2% increase in happiness isprobably influenced by many factors. We believe these resultsare impressive and suggest that small should not be equatedwith unimportant or disappointing. For instance, attachmenttheory has been a useful tool for researchers yet there isresearch showing that attachment (ex. categories, strength)explains only very small variance (as low as 1%) in variablessuch as delinquency (Cooper et al., 1998). These findings do notmean that attachment is not important. Similarly, although 2%might not be as high as some people might expect, friendshipstill adds something extra to our lives even though one mighthave a certain set point. For these reasons, we believe that thefindings pertinent to friendship quality are important.

The question remains though: which aspect of friendshipquality most strongly predicts happiness controlling for person-ality? Our analyses provided an answer: companionship and


self-validation. It might be surprising at first not to find inti-macy or help on this list since research established closeness andsupport as important correlates of well-being (Reis et al., 2000).The findings remind us of the famous argument put forth byWellman (1992), who suggested that considering friendship asonly being an intimate and supportive relationship limits ‘‘one�sworldview to a California hot tub (p. 104).’’ Apparently, atleast for happiness and for adults, one needs to consider otherfeatures of friendship as well. It might be that intimacy andhelp could be important for different indices of well-being, butnot for happiness and for this sample.

We evaluated our finding that companionship and self-validation were the most important dimensions predictinghappiness within the context of empirical research and theory.Companionship is one of the most basic dimensions of friend-ship. It refers to shared activities we engage in with friendsthat might ‘‘arouse enjoyment, amusement, and excitement(Mendelson and Aboud, 1999).’’ We want our friends to bewith us, for instance, when we go to a theater or a bar. Ofcourse, individuals differ in terms of what they do with theirfriends. Nevertheless, Americans rate companionship as themost important feature of friendship (Parlee, 1979). Moreover,in a series of studies, Rook (1987) compared the role of com-panionship and support in well-being and relationship satisfac-tion. She reported that companionship was more importantthan social support, especially during non-stressful times, inpredicting relationship satisfaction and loneliness. Likewise,Baldassare et al. (1984) found that companionship was the bestpredictor of happiness among the elderly.

Finding that companionship, doing things with a friend, wasone of the most important dimensions of friendship that pre-dicts happiness is also consistent with the model of Lyubomir-sky et al. (2005). It seems that what we do and with whom hasthe potential to change the set point of happiness(Csikszentmihalyi and Hunter, 2003). Even though the role ofcompanionship in happiness confirms previous research and fitsvery well with existing theories, one should be cautious. Thiscaution stems from the possible conceptual overlap between the


items used to assess companionship and happiness (especiallywith positive affect). We assessed companionship with five itemsand sample items were ‘‘my best friend makes me laugh’’ and‘‘my best friend is exciting to be with.’’ Nevertheless, the bivari-ate correlation between companionship and happiness was mod-erate (r = 0.28) and was only slightly stronger than thecorrelation between self-validation and happiness (r = 0.25).Although replications of this finding with affect-free compan-ionship scales is necessary to bolster confidence in the results,we find it safe to argue that companionship is an importantdimension of friendship for happiness among young adults.

The second feature of friendship that predicted happinessabove and beyond personality was self-validation. Self-validationas assessed in the present study was conceptualized as ‘‘perceivingthe other as reassuring, agreeing, encouraging, listening, andotherwise helping to maintain one�s self-image as a competentand worthwhile person’’ (Mendelson and Aboud, 1999) and wasassessed with items such as ‘‘points out things that I am good at’’and ‘‘compliments me when I do something well.’’ We believethat our finding could be explained by self-verification theory(Swann, 1990, Swann et al., 1992). Simply put, self-verification isa need whereby individuals seek to confirm their self-view, evenwhen it could be negative. Self-verification is a crucial process inthat individuals develop closer relationships with those who ver-ify their self-views (De La Ronde and Swan, 1998; McNulty andSwan, 1994; Swann et al., 1994) and self-verification is relatedto relationship satisfaction, depression and marital happiness(Giesler and Swann, 1999; Schafer et al., 1996). In a similar vein,Weisz and Wood (2005) found that receiving verification of thevalued identity and roles from a new friend predicted if the newfriend would be a best friend after 4 years. Considering thetheoretical link between self-verification and relationship andwell-being outcomes, it is safe to argue that hearing self-verifyingcomments from a friend (ex: you have a strong character, youdeal with problems very well) not only has the potential to in-crease closeness felt to the friend, but also, as this study showed,contributes to young adults� happiness above and beyond theinfluence of personality.


The findings of the present study highlight one importantpoint. Future studies investigating the relationship betweenfriendship and happiness should go beyond simply assessingnumber of friends or satisfaction with friends and assess quali-tative aspects of friendships. As the present study showed, dif-ferent features of friendship influence the happiness ofindividuals. It is important that future studies focus onqualitative features of friendship when studying friendship andhappiness.

Even though we found that friendship was still predictive ofhappiness above and beyond personality, one might wonder asto whether similar results might be obtained if personality wasassessed differently. For instance, McAdam�s (1985) power andintimacy motivation and Bakan�s (1966) agency and commu-nion constructs have been associated with well-being and rela-tionships (Helgeson, 1994; Zeldow et al., 1988). We suspect thatfriendship would still be predictive of well-being while control-ling for these personality variables. This expectation stems fromthe arguments that these constructs might be related to Big Fiveto some extent (Digman, 1997). We are aware that only futureresearch could shed light on this issue. Testing the importanceof friendship in well-being across different personality assess-ments could be an interesting line of research.

In the present study, we assessed happiness as it was com-monly assessed in the literature. Recent advances in the fieldconceptualized happiness with positivity ratios, rate of positiveto negative affect (Fredrickson and Losada, 2005). One mightwonder whether similar results would be obtained if onefocused on positivity ratios. Prior research has shown that posi-tive and negative affect was related to the friendship experiences(Beery et al., 2000) and our findings at the correlational level(Table I) were consistent with the literature. Also, Fredricksonand Losada (2005) reported that those individuals whose posi-tivity ratios exceeded 2.9 were found to experience positivesocial functioning (e.g., social acceptance). Considering theavailable research, we believe similar results would be obtainedif one relied on positivity ratios to assess happiness. We believethat investigation of whether friendship experiences would


predict happiness as assessed with positivity ratio might be afruitful line of research.

The finding that friendship quality was predictive of happi-ness above and beyond the influence of personality might raisesome interesting questions. Since marriage is considered to be a‘‘circumstance’’ in the model of Lyubumirsky et al. (2005), canlong-lasting friendship be considered a circumstance as well?Also, would long-lasting friendships explain more variance inhappiness than friendships formed and experienced duringcollege?

First of all, we believe that one should distinguish betweenhaving a life-long friendship and the quality of that friendship.One might have a friendship that lasted several years but simplyknowing this might not be optimal to understand the role offriendship in happiness because friendship is a qualitative rela-tionship. Considering this, we believe that friendship, regardlessof the duration, should be considered as an activity rather thana circumstance. Second of all, how common is long-lastingfriendships? Most of the research in this area is done with olderadults and documented that older adults have friendships thatlasts as high as 30 years (Blieszner, 1989; Litwak, 1989;Matthews, 1986; Parlee, 1979; Roberto and Kimboko, 1989;Shea et al., 1988; Wenger and Jerrome, 1999). Even thoughindividuals do not live in close proximity to their life-longfriends (Litwak, 1989; Matthews, 1986), adults who had long-lasting friendships were willing to take long trips to meet theirfriends (Matthews, 1986); were still exchanging resources andaffection and were expecting to carry the same level of intimacyinto the future (Blieszner 1989); and were exchanging adviceand personal information more often in old friendships than innew friendships. These findings suggest that the overall qualityof long-lasting friendships might be high in quality.

Considering these points, then, would long-lasting friendshipexplain more variance in happiness when compared againststrong dispositional variables? We would predict that, control-ling for the frequency of contact, the answer to this questionwould be yes. First of all, the overall quality of friendship expe-rience might be higher in these friendships compared to those


newly formed ones. Second of all, long-lasting friendships havethe potential to become an attachment bond (Fraley and Davis,1997). Also, recent research suggests that friendships that couldbe classified as attachment bonds (e.g., using the other for asecure base) are higher in overall quality and contribute to hap-piness significantly more when compared to friendships that didnot qualify for an attachment bond (Demir et al., 2006a). Con-sidering these points, we believe that long lasting friendships,especially those becoming an attachment bond, might explainmore variance happiness above and beyond the influence ofpersonality. Future research on this point might enhanceour understanding of the association between friendship andhappiness.

Despite the interesting and theoretically important findings,this study was not without limitations. First of all, the presentstudy was cross-sectional in nature. Throughout the paper wetreated friendship as a predictor of happiness; however, it mightbe that being happy sets the condition for experiencing highquality friendships. Longitudinal designs have the potential todocument a clearer picture. Nevertheless, our findings were con-sistent with theoretical arguments and empirical research.Second of all, the findings for friendship should be interpretedwith caution for it might not be generalizable to young adultswho are not in college and other age groups. Friendships arereported to be important across the life-span (Hartup andStevens, 1997); however, we do not know if friendship qualitywould predict happiness above and beyond the influence ofpersonality among other age groups. Furthermore, college stu-dents, as compared to other age groups, spend more time withfriends in general (Carbery and Buhrmester, 1997) and mightvalue companionship more than other features. That is, otherfeatures of friendship might emerge as important predictors ofhappiness among other age groups. These points highlight theneed of further research in this area to increase confidence inthe results reported.

Another limitation of the present study was the ratio ofwomen to men in the sample. There were more women thanmen in the analyses reported. Studies relying on college students


as their participants might inevitably face such a problem. Thisraises the following question: Had equal number of men andwomen participated in the study, would the results have beendifferent? We do not have an answer to this question at thispoint and as such it is the task of future research to investigatethis issue. Finally, although our focus was on happiness, somemight argue that other dimensions of well-being should be con-sidered as well (e.g., depression, self-esteem), since predictors ofhappiness and other well-being indices might be different.

The present study investigated the role friendship plays inhappiness. Our results showed that friendship quality contrib-uted to happiness above and beyond the influence of genderand personality. This finding gave support to the theoreticalarguments and showed that friends add something extra to ourlives. Further analyses revealed that it was companionship andself-validation that contributed to happiness while controllingfor personality.

ACKNOWLEDGEMENT

The authors would like to thank Douglas Barnett for hissupport and assistance throughout this research project.

REFERENCES

Argyle, M.: 2001, The Psychology of Happiness (Routledge, London).Argyle, M. and L. Lu: 1990, �Happiness and social skills�, Personality andIndividual Differences 11, pp. 1255–1261.

Asendorpf, J.B. and S. Wilpers: 1998, �Personality effects on social relation-ships�, Journal of Personality and Social Psychology 74, pp. 1531–1544.

Bakan, D.: 1966, The Duality of Human Existence (Rand McNally, Chicago).Baldassare, M., S. Rosenfield and K.S. Rook: 1984, �The types of social rela-tions predicting elderly well-being�, Research on Aging 6, pp. 549–559.

Baumeister, R.F. and M.R. Leary: 1995, �The need to belong: Desire forinterpersonal attachment as a fundamental human motivation�, Psycholog-ical Bulletin 117, pp. 497–529.

Berry, D.S., J.K. Willingham and C.A. Thayer: 2000, �Affect and personality aspredictors of conflict and closeness in young adults� friendships�, Journal ofResearch in Personality 34, pp. 84–107.


Blieszner, R.: 1989, �Developmental processes of friendship�, in R.G. Adamsand R. Blieszner (eds.), Older Adult Friendship: Structure and Process(Sage, London), pp. 108–126.

Bukowski, W.M, M. Boivin and B. Hoza: 1994, �Measuring friendship qualityduring pre- and early adolescence: The development and psychometricproperties of the Friendship Qualities Scale�, Journal of Social and PersonalRelationships 11, pp. 471–484.

Bukowski, W.M. and B. Hoza: 1989, �Popularity and friendship: Issues intheory, measurement, and outcome�, in T.J. Berndt and G.W. Ladd(eds.), Peer Relationships in Child Development (Wiley, New York),pp. 15–45.

Burt, R.S.: 1987, �A note on strangers, friends, and happiness�, Social Networks9, pp. 311–331.

Carbery, J. and D. Buhrmester: 1997, �Friendship and need fulfillment duringthree phases of young adulthood�, Journal of Social and Personal Rela-tionships 15, pp. 393–409.

Cheng, H. and A. Furnham: 2001, �Attributional style and personality aspredictors of happiness and mental health�, Journal of Happiness Studies 2,pp. 307–327.

Compton, W.C.: 1998, �Measures of mental health and a five factor theory ofpersonality�, Psychological Reports 83, pp. 371–381.

Cooper, M.L., P.R. Shaver and N.L. Collins: 1998, �Attachment styles, emo-tion regulation, and adjustment in adolescence�, Journal of Personality andSocial Psychology 74, pp. 1380–1397.

Csikszentmihalyi, M. and J. Hunter: 2003, �Happiness in everyday life: Theuses of experience sampling�, Journal of Happiness Studies 4, pp. 185–199.

Davis, K.E. and M.J. Todd: 1985, �Assessing friendship: Prototypes, paradigmcases and relationship description�, in S. Duck and D. Perlman (eds.),Understanding Personal Relationships: An interdisciplinary approach (Sage,London), pp. 17–38.

De La Ronde, C. and W.B. Swann, Jr.: 1998, �Partner verification: Restoringshattered images of our intimates�, Journal of Personality and SocialPsychology 75, pp. 374–382.

Demir, M., M. Ozdemir and L.A. Weitekamp: 2006a, Attachment networksamong young adults: Factors predicting strength of attachment and impli-cations of full-blown attachments (submitted).

Demir, M. and K.A. Urberg: 2004, �Friendship and adjustment amongadolescents�, Journal of Experimental Child Psychology 88, pp. 68–82.

Demir, M., L.A. Weitekamp and M. Ozdemir: 2006b, �Looking to happytomorrows friends: Best and close friendships as they predict happiness�,Journal of Happiness Studies (in press).

DeNeve, K.M. and H. Cooper: 1998, �The happy personality: A meta-analysisof 137 personality traits and subjective well-being�, Psychological Bulletin124, pp. 197–229.

Diener, E.: 1984, �Subjective well-being�, Psychological Bulletin 95, pp. 542–575.


Diener, E.: 1994, �Assessing subjective well-being: Progress and opportunities�,Social Indicators Research 31, pp. 103–157.

Diener, E., R.A. Emmons, R.J. Larsen and S. Griffin: 1985a, �The satisfactionwith life scale�, Journal of Personality Assessment 49, pp. 71–75.

Diener, E. and R. Lucas: 1999, �Personality, and subjective well-being�, inD.Kahneman, E.Diener andN. Schwarz (eds.),Well-being: TheFoundationsof Hedonic Psychology (Russell Sage Foundation, New York), pp. 213–229.

Diener, E., R.E. Lucas and C.N. Scollon: 2006, �Beyond the hedonic treadmill:Revising the adaptation theory of well-being�, American Psychologist 61,pp. 305–314.

Diener, E. and M.E.P. Seligman: 2002, �Very happy people�, PsychologicalScience 13, pp. 81–84.

Diener, E., E.M. Suh, R.E. Lucas and H.L. Smith: 1999, �Subjective well-being:Three decades of progress�, Psychological Bulletin 125, pp. 276–302.

Digman, J.M.: 1997, �Higher-order factors of the Big Five�, Journal ofPersonality and Social Psychology 73, pp. 1246–1256.

Fehr, B.: 1996, Friendship Processes (SAGE Publications, London).Fraley, R.C. and K.E. Davis: 1997, �Attachment formation and transfer inyoung adults� close friendships and romantic relationships�, Personal Rela-tionships 4, pp. 131–144.

Fredrickson, B.L. and M.F. Losada: 2005, �Positive affect and the complexdynamics of human flourishing�, American Psychologist 61, pp. 678–686.

Furnham, A. and H. Cheng: 1997, �Personality and happiness�, PsychologicalReports 80, pp. 761–762.

Giesler, R.B. and W.B. Swann, Jr.: 1999, �Striving for confirmation: The role ofself- verification in depression�, in T.E. Joiner and J.C. Coyne (eds.), TheInteractional Nature of Depression: Advances in Interpersonal Approaches(APA Press, Washington, D.C.), pp. 189–217.

Gladow, N.W. and M.P. Ray: 1986, �The impact of informal support systemson the well-being of low income single parents�, Family Relations: Journal ofApplied Family and Child Studies 35, pp. 113–123.

Hartup, W.W. and N. Stevens: 1997, �Friendship and adaptation in the lifecourse�, Psychologocial Bulletin 121, pp. 355–370.

Hays, R.B.: 1988, �Friendship�, in S. Duck (ed.), Handbook of PersonalRelationships: Theory, Research, and Interventions (Wiley, New York),pp. 391–408.

Helgeson, V.S.: 1994, �Relation of agency and communion to well-being:Evidence and potential explanations�, Psychological Bulletin 116, pp. 412–428.

Hills, P. and M. Argyle: 2001, �Happiness, introversion–extroversion andhappy introverts�, Personality and Individual Differences 30, pp. 595–608.

Hinde, R.A.: 1997, Relationships: A Dialectical Perspective (Psychology Press,Taylor & Francis Ltd).

Jensen-Campbell, L.A., R. Adams, D.G. Pery, K.A. Workman, J.Q. Furdellaand S.K. Egan: 2002, �Agreeableness, extraversion, and peer relations in


early adolescence: Winning friends and deflecting aggression�, Journal ofResearch in Personality 36, pp. 224–251.

John, O. P. and S. Srivastava: 1999, �The Big Five trait taxonomy: History,measurement, and theoretical perspectives�, in L.A. Pervin and O.P. John(eds.), Handbook of Personality: Theory and research (Guilford, NewYork),pp. 102–138.

Kasser, T. and K.M. Sheldon: 2002, �What makes for a merry Christmas?�,Journal of Happiness Studies 3, pp. 313–329.

Litwak, E.: 1989, �Forms of friendships among older people in an indsustrialsociety�, in R.G. Adams and R. Blieszner (eds.), Older Adult Friendship:Structure and Process (Sage, London), pp. 45–63.

Lu, L.: 1999, �Personal or environmental causes of happiness: A longitudinalanalysis�, Journal of Social Psychology 139, pp. 79–90.

Lucas, R.E. and E. Diener: 2001, �Understanding extraverts� enjoyment ofsocial situations: The importance of pleasantness�, Journal of Personalityand Social Psychology 81, pp. 343–356.

Lucas, R.E., E. Diener, A. Grob, E.M. Suh and L. Shao: 2000, �Cross-culturalevidence for the fundamental features of extraversion�, Journal of Person-ality and Social Psychology 79, pp. 452–468.

Lyubomirsky, S., K.M. Sheldon and D. Schkade: 2005, �Pursuing happiness:The architecture of sustainable change�, Review of General Psychology 9,pp. 111–131.

Lyubomirsky, S., C. Tkach and M.R. DiMatteo: 2006, �What are thedifferences between happiness and self-esteem?�, Social Indicators Research78, pp. 363–404.

Mathur, R.: 2004(March), Relations between friendship quality and measuresof well-being, Paper presented at the biennial meeting of the Society forResearch on Adolescence, Baltimore, MD, USA.

Matthews, S.H.: 1986, Friendships Trough the Life Course: Oral Biographiesin Old Age (Sage, Beverly Hills).

McAdams, D.P.: 1985, Power, Intimacy, and the Life Story: PsychologicalInquiries into Identity (Dow-Jones-Irwin, Homewood, IL).

McCrae, R.R. and P.T. Costa: 1990, Personality in Adulthood (Guilford Press,New York).

McCrae, R.R. and P.T. Costa: 1991, �Adding liebe and arbeit: The full five-factor model and well-being�, Personality and Social Psychology Bulletin 17,pp. 227–232.

McNulty, S.E. and W.B. Swann: 1994, �Identity negotiation in roommaterelationships: The self as architect and consequence of social reality�, Journalof Personality and Social Psychology 67, pp. 1012–1023.

Mendelson, M.J. and F.E. Aboud: 1999, �Measuring friendship quality in lateadolescents and young adults: McGill Friendship Questionnaires�, CanadianJournal of Behavioural Science 31, pp. 130–132.

Morrison, K.A.: 1997, �Personality correlates of the Five-Factor Model for asample of business owners/managers: Associations with scores on self-


monitoring, type A behavior, locus of control, and subjective well-being�,Psychological Reports 80, pp. 255–272.

Myers, D.: 2000, �The funds, friends and faith of happy people�, AmericanPsychologist 55, pp. 56–67.

Parker, J.G. and S.R. Asher: 1993, �Friendship and friendship quality in middlechildhood: Links with peer group acceptance and feelings of loneliness andsocial dissatisfaction�, Developmental Psychology 29, pp. 611–621.

Parlee, M.B.: 1979, �The friendship bond�, Psychology Today 13, pp. 43–54,113.

Reis, H.T., W.A. Collins and E. Berscheid: 2000, �The relationship context ofhuman behavior and development�, Psychological Bulletin 126, pp. 844–872.

Roberto, K.A. and P.J. Kimboko: 1989, �Friendship in later life: Definitionsand maintenance patterns�, International Journal of Aging and HumanDevelopment 28, pp. 9–19.

Rook, K.S.: 1987, �Social support versus companionship: Effects on life stress,loneliness, and evaluations by others�, Journal of Personality and SocialPsychology 52, pp. 1132–1147.

Ruiz, W.M.: 2005, �The five-factor model of personality, subjective well-being,and social adaptation: Generalizability to the Spanish context�, Psycholog-ical Reports 96, pp. 863–866.

Schafer, R.B., K.A.S. Wickrama and P.M. Keith: 1996, �Self-concept discon-firmation, psychological distress, and marital happiness�, Journal ofMarriage and the Family 58, pp. 167–177.

Seligman, M.E.P., T.A. Steen, N. Park and C. Peterson: 2005, �Positivepsychology progress: Empirical validation of interventions�, AmericanPsychologist 60, pp. 410–421.

Shea, L., L. Thompson and R. Blieszner: 1988, �Resources in older adults�old and new friendships�, Journal of Social and Personal Relationships 5,pp. 83–96.

Sheldon, K.M. and T. Kasser: 2001, �Getting older, getting better? Personalstrivings and personality development across the life-course�, DevelopmentalPsychology 37, pp. 491–501.

Sheldon, K.M., T. Kasser, L. Houser-Marko, T. Jones and D. Turban: 2005,�Doing one�s duty: Chronological age, felt autonomy, and subjectivewell-being�, European Journal of Personality 19, pp. 97–115.

Sheldon, K.M. and S. Lyubomirsky: 2006b, �Achieving sustainable gains inhappiness: Change your actions, not your circumstances�, Journal of Hap-piness Studies 7, pp. 55–86.

Sheldon, K. and S. Lyubomirsky: 2004, �Achieving sustainable new happiness:Prospects, practices, and prescriptions�, in P.A. Linley and S. Joseph (eds.),Positive Psychology in Practice (Wiley, Hoboken, NJ), pp. 127–145.

Swann, W.B., Jr.: 1990, �To be adored or to be known? The interplay of self-enhancement and self-verification�, in R.M. Sorrentino and E.T. Higgins(eds.), Motivation and Cognition (Guilford Press, New York), pp. 404–448.


Swann, W.B., Jr., C. De La Ronde and J.G. Hixon: 1994, �Authenticity andpositivity strivings in marriage and courtship�, Journal of Personality andSocial Psychology 66, pp. 857–869.

Swann, W.B., Jr., A. Stein-Seroussi and R.B. Giesler: 1992, �Why people self-verify�, Journal of Personality and Social Psychology 62, pp. 392–401.

Tellegen, A., D.T. Lykken, T.J. Bouchard, K.J. Wilcox, N.L. Segal andS. Rich: 1988, �Personality similarity in twins reared apart and together�,Journal of Personality and Social Psychology 54, pp. 1031–1039.

Watson, D., L.A. Clark and A. Tellegen: 1988, �Development and validation ofbrief measures of positive and negative affect: The PANAS scales�, Journalof Personality and Social Psychology 54, pp. 1063–1070.

Weisz, C. and L.F. Wood: 2005, �Social identity support and friendship out-comes: A longitudinal study predicting who will be friends and best friends4 years later�, Journal of Social and Personal Relationships 22, pp. 416–432.

Wellman, B.: 1992, �Men in networks: Private communities, domestic friend-ships�, in P.M. Nardi (ed.), Men�s Friendships (Sage, Newbury Park, CA),pp. 74–114.

Wenger, G.C. and D. Jerrome: 1999, �Change and stability in confidant rela-tionships: Findings from the Bangor longitudinal study of ageing�, Journalof Aging Studies 13, pp. 269–294.

Wilkinson, R.B. and W.A. Walford: 2001, �Attachment and personality in thepsychological health of adolescents�, Personality and Individual Differences31, pp. 473–484.

Wright, P.H.: 1985, �The acquaintances description form�, in S. Duck andD. Perlman (eds.), Understanding Personal Relationships: An Interdisci-plinary Approach (Sage, London), pp. 39–62.

Zeldow, P.B., S.R. Daugherty and D.P. McAdams: 1988, �Intimacy, power,and psychological well-being in medical students�, Journal of Nervous andMental Disease 176, pp. 182–187.

Address for correspondence:MEL_IKSAH DEM_IRDepartment of PsychologyWayne State UniversityDetroit, MI, 48201USA

E-mail: [email protected]


Happiness n Friendship

Documents

de la ronde

point scale

recent theoretical

incremental

older adult

understanding

negative affect

measuring