Intimate friendship relationships between young adults

Abstracts

The literature is consensual regarding the importance of friendships throughout the lifecycle; most of the time for improvements in the quality of life. The aim of this study was to investigate and describe intimate friendship relationships between young adults. Participants were 98 young adults aged between 18 and 30. Sampling procedures were based on the Respondent Driven Sampling technique. The instrument was composed of three self-report questionnaires and data were analyzed quantitatively. Results showed homogeneity of characteristics among intimate friends, especially concerning gender. Companionship emerged as the most distinctive aspect of friendships. All friendship functions showed positive correlations with each other. There seems to be a filter of similarities in the friendships of young adults. Also, friendships that show good quality in one specific function tend to be good quality friendships as a whole.

friendship; interpersonal relations; young adults; adults

A literatura é consensual quanto ao importante papel dos amigos no ciclo vital, na maior parte do tempo, para melhorias na qualidade de vida. Este estudo teve como objetivo investigar e descrever características dos relacionamentos íntimos de amizade de jovens adultos. Participaram 98 jovens adultos com idade entre 18 e 30 anos. Os procedimentos de amostragem foram baseados na técnica do Respondent Driven Sampling. Foram utilizados três questionários autoaplicados e os dados foram submetidos a tratamento quantitativo. Os resultados demonstraram uma homogeneidade de características entre amigos íntimos, especialmente para o gênero. O companheirismo despontou como aspecto mais marcante na amizade. Todas as funções da amizade apresentaram correlações positivas entre si. Conclui-se que parece haver um filtro de similaridades entre amigos na adultez jovem e que amizades de boa qualidade em uma determinada função costumam ser de boa qualidade em seu total.

amizade; relações interpessoais; jovens; adultos

ARTICLE

Intimate friendship relationships among young adults

Relaciones de amistad íntima entre adultos jóvenes

Diogo Araújo De Sousa^I; Elder Cerqueira-Santos^II

^IUniversidade Federal do Rio Grande do Sul, Porto Alegre-RS, Brazil

^IIUniversidade Federal de Sergipe, São Cristóvão-SE, Brazil

^{Correspondence to}

ABSTRACT

The literature is consensual regarding the importance of friendships throughout the lifespan; most of the time for improvements in the quality of life. The aim of this study was to investigate and describe intimate friendship relationships among young adults. Participants were 98 young adults aged between 18 and 30. Sampling procedures were based on the Respondent Driven Sampling technique. The instrument was composed of three self-report questionnaires and data were analyzed quantitatively. Results showed homogeneity of characteristics among intimate friends, especially concerning the gender. Companionship emerged as the most distinctive aspect of friendships. All friendship functions showed positive correlations with each other. There seems to be a filter of similarities in the friendships of young adults. Also, friendships that show good quality in one specific function tend to be good quality friendships as a whole.

Keywords: friendship, interpersonal relations, young adults, adults

RESUMEN

La literatura es consensual cuanto a la importancia de la amistad a lo largo del ciclo vital; en la mayor parte del tiempo, para la mejoría de la calidad de vida. Este estudio tuvo como objetivo investigar y describir las relaciones íntimas de amistad de adultos jóvenes. Participaron 98 adultos jóvenes de edades comprendidas entre 18 y 30 años. El muestreo estuvo basado en la técnica del Respondent Driven Sampling. Fueron utilizados tres cuestionarios auto-administrados y los datos fueron analizados cuantitativamente. Los resultados demostraron homogeneidad en las características entre amigos íntimos, especialmente para el género. El compañerismo se destacó como el aspecto más importante de la amistad. Todas las funciones de la amistad mostraron correlaciones positivas entre sí. Se concluye que hay un filtro de similitudes entre amigos en la adultez joven y que amistades que son buenas en un aspecto tienden a ser buenas amistades en su total.

Palabras clave: amistad, relaciones interpersonales, jóvenes, adultos

Friendship is a form of relationship that virtually everyone experiences with different people throughout life. Friendship relationships are a complex and multifaceted phenomenon with various definitions in the literature (Bukowski, Newcomb, & Hartup, 1996; Bukowski, Motzoi, & Meyer, 2009; Souza & Hutz, 2007). The difficulty in defining friendship lies not only in its multifaceted character, but also in the fact that the theories about interpersonal relationships tend to target family and romantic relationships. Friendship relationships remain "the least investigated of the three areas, requiring a greater number of theoretical and empirical studies" (Garcia, 2005, p. 285th). There is a lack of more robust, theoretically and empirically based theories specifically consolidated for friendship (Garcia, 2005; Souza & Hutz, 2008a).

Friendships are found in the majority of, if not all, societies, and generally during the various phases of life. Although they manifest differently depending on the cultural system, friendships always have some aspects present, no matter where or at what moment of the lifecycle they are established. For example, it is a voluntary relationship that involves mutual appreciation and reciprocity. Friends have interests, tastes or other characteristics in common. Furthermore, friends help and are committed to one another, presenting higher levels of cooperation than non-friends. Thus, friendship can be characterized as an intimate bilateral, mutual and voluntary relationship (Bukowski et al., 2009; Krappmann, 1996; Lisboa & Koller, 2003).

Asher, Parker and Walker (1996) describe some requirements consensually identified as important for the formation and maintenance of a friendship. The most important requirements being: (a) willingness to invest free time in the friendship, (b) companionship, (c) reciprocity, and (d) intimacy. In addition to these four, other equally important aspects are the adequate expression of care, concern, admiration and affection, help, advice, comfort and emotional support, demonstration of trust, loyalty and conflict resolution strategies. Fehr (1996) constructed a compilation of definitions of friendship widely used in the field of Developmental Psychology and, from this, conceptualized friendship as "a personal and voluntary relationship, which provides intimacy and help, in which both parties like each another and seek each other's company" (p. 7).

Friendship is a significant relationship for people, which promotes happiness and life satisfaction through instrumental rewards, emotional support and companionship (Argyle, 2001). Mendelson and Aboud (1999) identified and categorized six different component functions of friendship: (a) stimulating companionship, (b) help, (c) intimacy, (d) reliable alliance, (e) self-validation, and (f) emotional security. The function of stimulating companionship takes into account engaging in enjoyable, entertaining and stimulating activities together. Help addresses the provision of guidance, advice, assistance and other forms of support. Intimacy concerns sensitivity to the states and needs of the other, providing openness to honest expression of thoughts, feelings and personal information. Reliable alliance reflects availability and continuous loyalty. Self-validation involves the function of reassuring, encouraging and helping each other to maintain a positive self-image. Finally, emotional security covers the provision of comfort and trust in new or threatening situations.

Analyzing the definitions and characterizations present in the literature, the typical aspects of friendship relationships can be categorized into three spheres. First, a sphere of a social nature, in which characteristics such as companionship, loyalty and commitment are included. Second, a sphere of an instrumental character, in which functions of help, advice and exchanges between friends are found. Third, a sphere of an affective character, involving intimacy, mutual appreciation and the loving characteristics of friendship relationships.

The developmental process of a friendship relationship depends on the convergence of different factors: (a) environmental - for example, residential proximity, places where they spend the day and communication aspects in the social network; (b) situational - such as interaction, frequency of contact and availability; (c) individual - covering the inclusion and exclusion criteria used to elect someone as a friend, and (e) dyadic - among which mutual appreciation and self-disclosure stand out (Fehr, 1996). Therefore, friendships are subject to constant change throughout the lifecycle, caused by changes in individual or dyadic-interactional aspects, such as by different situational or environmental configurations (Souza & Hutz, 2008b).

Once the relationship has been established, the differences between the types and levels of friendship are determined by increments in the characteristics associated with it. For example, the questions of intimacy, support and self-disclosure tend to evolve to the extent to which these friends become closer. Therefore, Souza and Hutz (2008b) argue that the differences between friendships are often more quantitative than qualitative. Thus, in a higher level of friendship - between those called close friends or best friends - a proportionately greater degree of acceptance, support and intimacy is found.

Friendship relationships have the function of promoting affection, intimacy and trust. In childhood, the relationship with friends helps the healthy socialization process, enabling them to experience and deal with positive relational aspects, such as cooperation and social support; and negative ones, such as conflict and competition (Daudt, Souza, & Sperb, 2007, Garcia, 2005). For Lisboa and Koller (2003), the experience of love and affection originating from the friendship relationships is unique in the lifecycle. The authors argue, for example, that the relationships with parents carry a social expectation that parents should love their children, which makes the affection and acceptance that a child experiences in the family relationships different to that demonstrated freely by a friend.

In adolescence, friendship relationships provide the main source of intimacy for youths (Pereira & Garcia, 2007). More than with the family, teenagers spend most of their time with other young people, and use much of their leisure time in socializing activities with friends and peer groups (Oliveira, Camilo, & Assunção, 2003). Generally, in childhood and adolescence the friendship relationships promote social engagement, cooperation and conflict management. Furthermore, between friends, self-concept and self-esteem are highlighted, as well as knowledge about others and about what surrounds them. Friendships serve as both emotional and cognitive developmental resources that, among other benefits, protect the youths while passing through negative experiences (Adams, Santo, & Bukowski, 2011; Bukowski et al., 1996).

In the scientific literature, studies of child and adolescent friendship have received more attention; with young adults and adults being less investigated, especially in Brazil (Duarte & Souza, 2010; Souza & Hutz, 2008a, 2008b). Perceiving the lack of studies on friendship in adulthood, Souza and Hutz (2008a, 2008b) investigated this form of relationship specifically in this population. The authors contend that, during this phase of the lifecycle, friendship is characterized by homogeneity in various aspects, such as gender, age, marital status, education, occupational status, income, religion, ethnicity, personality traits, interests and shared activities. Furthermore, friendships in early adulthood narrow and strengthen to the extent that young people move away from their family members (Peron, Guimarães, & Souza, 2010). The number of friends appears to be at its peak in the first years of young adulthood, reducing throughout the passage to intermediate adulthood (DeSousa & Cerqueira-Santos, 2011).

Other recent Brazilian studies had also been dedicated to research into friendship relationships in populations over 18 years of age. Duarte and Souza (2010) conducted a study of university students aged between 18 and 30 years, in which some characteristics relevant to friendship for people in this age group were identified. Among these characteristics positive aspects are highlighted, such as trust, shared intimacy, respect and acceptance, and also negative aspects, such as conflicts and submission. Gomes and Silva Júnior (2007) followed workers of popular cooperatives, over 20 years of age, investigating their friendship relationships. The authors discuss how solidarity and commitment between friends provided acceptance, strength, care and support among the members of the cooperatives, strengthening the bonds and the engagement in the community.

It is of great interest and relevance to study the way the friendship relationships are constituted in the different stages of the lifecycle. As discussed, friends play an important role in improving the quality of life of people. For example, the friendship relationships continue to be major providers of social support in adulthood. Such support is an important coping resource, helping people to deal with impacting diseases such as cancer (Santana, Zanin, & Maniglia, 2008) and the coronary heart diseases (Abreu-Rodrigues & Seidl, 2008), as well as to face crises in general through the provision of emotional and practical/instrumental support (DeSousa & Cerqueira-Santos, 2012).

We agree with Souza and Hutz (2007) regarding the fact that friendship relationships "deserve attention with the study of their processes and dimensions, and the investigation of the perception of quality through their related functions, satisfaction and feelings" (p. 94). Thus, the aim of the present study was to investigate and describe characteristics of the intimate friendship relationships of young adults. Young adulthood was comprehended as the period between the end of adolescence and the beginning of the 30s (Rawlins, 1992).

Method

Participants

A total of 124 young adults, aged between 18 and 30 years, were recruited to participate in the study. Data collection began with four seeds - two men and two women, for comparison purposes and to maintain an initial equal distribution. Through the Respondent Driven Sampling (RDS) technique the 124 likely participants were contacted, representing the sample size after reaching the fifth wave in the research design. However, 26 people contacted were unable to participate or did not return the questionnaires by the established deadline. The final sample therefore consisted of 98 participants (45.9% men).

As explained by Goel and Salganik (2009), RDS is a new sampling approach, designed for use with difficult to access populations, which has achieved great international popularity. Neiva-Silva (2010) describes how this method uses innovative technologies from the principles of the theory of Markov. The data are collected through a snowball mechanism, in which sample members recruit new participants themselves. The study begins with the recruitment of a few people who fit the profile of the target population - the seeds. After their participation, each seed is asked to indicate people known to them that fit the target profile for participation. The sampling continues in this way until the desired sample size is reached. The seeds are called the 1^st wave; those recruited by them constitute the 2^nd wave, and so on. This process results in a recruitment network with characteristics that negate possible bias present in the initial choice of seeds (Goel & Salganik, 2009).

Although the target population of this study was not a difficult access population, the design based on RDS proved useful because, as summarized by Neiva-Silva (2010), it produces a final sample independent of those who initiated it and provides good quality information quickly. Furthermore, such a technique considers the social network characteristic of the sample. There is a gap in the studies on friendship relationships regarding this aspect (Bukowski et al., 1996).

Instrument

The instrument was developed based on that used by Souza and Hutz (2008a), with some alterations made. In the first part the respondents are asked what three words first come to mind when thinking about friendship and what friendship is for them. It also asks the participants to indicate close friendships (defined as friendships that stand out from the others and limited to 12 nominations), and then to specify how many among them are male and female, and how many actually live in the same city as them. Next they are asked to indicate one better friendship, about which some specific information is requested: (a) gender, (b) length of friendship, (c) where they met, (d) whether they reside in the same city, (e) whether there is another form of current relationship with them, (f) whether the person indicated knows that they are the best friendship of the participant, and (g) the weekly frequency of personal, telephone, email and internet contact.

The second part is composed of the McGill questionnaires, adapted to Brazil by Souza and Hutz (2007) for use with adult populations. The McGill questionnaires are used to evaluate the quality of the friendship according to the categorization described by Mendelson and Aboud (1999). The Brazilian version presented good psychometric properties in its adaptation study. The factorial structure found was similar to the original, with six factors related to the six functions theoretically underlying the construct of friendship in the instrument. This solution explained 57.6% of the variance in the study, and all six factors presented good levels of internal consistency (Cronbach's alphas between 0.73 and 0.81). Two unifactorial scales compose the questionnaires: one regarding the positive feelings (Cronbach's alpha = 0.71) and the other concerning the negative feelings (Cronbach's alpha = 0.84) associated with the best friendship (Souza & Hutz, 2007). Finally, the third part of the instrument questions the participant about the existence of other best friends (limited to five nominations) and the gender of these, and gathers bio-socio-demographic information about the respondents.

Procedure

Data collection. The instrument was self-applied and its application was carried out individually, at a place and time previously agreed with the participant. The four chosen seeds (1^st wave) answered the instrument and each was asked to indicate two other people to participate in the study, preferably from those indicated in the questionnaire as close friends. This procedure aimed to maintain the social network character promoted by RDS. The eight people that the seeds indicated (2^nd wave) were contacted to schedule a time and place for the application of the instrument, if they consented to participate.

Following the same procedure, each of the new respondents indicated two more friends to participate in the study, adding 16 components for the 3^rd wave, 32 for the 4^th wave and 64 for the 5^th and final wave. From the first contact with the participants, they were guaranteed comprehension of the study characteristics and of their rights as a respondent, including the voluntary nature of the participation and the confidentiality of the information. The signing of the Terms of Free Prior Informed Consent (TFPIC) was a requirement for participation. In the cases where it was indicated that the friend did not reside in the city of the study, was traveling or when personal contact with them was not possible, the application of the instrument was conducted via e-mail, after contact by telephone for the appropriate presentation and recruitment for participation in the study. In these cases, the TFPIC was sent to the participant to sign and digitized for return or, where this procedure was not possible, consent was obtained via email.

Data analysis. The data were subjected to quantitative treatment. Descriptive statistical analyzes were performed, through the survey of the frequencies, means and standard deviations, as well as bivariate analyses (chi-square, Student's t test and Pearson's correlation). Descriptive graphical representations of certain characteristics of the sample were also developed. The purpose of these graphical representations was to enable the visual analysis of the distribution of these characteristics in the social networks formed from the RDS sampling.

Ethical Considerations

The study was approved by the Human Research Ethics Committee of the Universidade Federal de Sergipe (protocol approval number 338/2010).

Results

The ages of the participants ranged from 18 to 30 years (M = 23.01, SD = 6.57). Regarding education, the majority had incomplete higher education (75.5%). In line with the more frequent level of education, most of the sample reported the occupation of student or trainee (67.3%). The remaining participants reported various professions. Regarding religion, 46.9% of the participants said they were Catholic. Concerning the relationship status, 53% were single and the remaining 47% were involved in some form of stable relationship, either dating, engaged or married. For the committed participants, the mean length of the relationship was 31.6 months (SD = 30.13 months).

There were significant differences in how the men and women were distributed in the social network of friends indicated by the participants. Men usually indicated more men as close friends and the women normally indicated more women. In terms of frequency, 74.1% of the men were nominated by other men and 22.2% by the women - the remaining 3.7% were the two male seeds. Similarly, 71.4% of the women were nominated by other women and 25.7% by men - the other 2.9% were the two female seeds, χ² = 30.07, df = 2, p < 0.001.

Analyzing the configuration of the social networks according to the age of the participants, a tendency to nominate people of a similar age as close friends can be noticed between the sampling waves. Groups of younger people and older people were formed in the networks. An ANOVA calculation was performed for comparison of the mean ages of the participants originating from each seed of the study, and the differences encountered between groups were significant, F_(3.94) = 17.60; p < 0.001. There were differences between seed A and seeds B and D, between seed A and seed C, and between seeds B and D and seed C. Seed A generated a network with a mean age of 25.59 years (SD = 2.00), the highest mean among the four seeds. Seed D generated a mean of 24.08 years (SD = 3.37), which did not differ significantly from the mean of seed B of 23.15 years (SD = 2.05). Finally, the lowest mean, 20.53 years (SD = 2.19), was generated by seed C.

Conversely, considering the relationship status of the participants, there were no significant differences between the indications of those who said they were single and of those who said they were involved in a steady relationship - dating, engaged or married. In terms of frequency, 55.8% of the single people were nominated by other single people and 40.4% by committed people - the remaining 3.8% were two seeds that were single. In contrast, 43.5% of the committed people were nominated by other committed people and 52.2% by single people - the other 4.3% were the two committed seeds. These differences did not achieve statistical significance, χ² = 0.13, df = 2, p = 0.937.

In relation to the first three words that came to mind when they thought of friendship relationships, the majority of the participants (65.3%) cited "companionship" and words related to its semantic field (e.g., complicity, partnership, fraternity and brotherhood). In second place came "trust" (44.9%), followed by "sincerity" (29.6%) and "respect" (24.5%). Other frequently cited words were "loyalty/commitment" (22.4%), "joy/fun/happiness" (19.4%), "Love" (17.3%), "affection/tenderness" (15.3 %) and "support/help" (15.3%).

The mean number of close friendships of the participants was 7.82 (SD = 2.85), with a minimum of two and maximum of 12 friends (the limit given in the questionnaire). The mean number of these friendships that were highlighted as living in the same city as the participant was 5.88 (SD = 3.21).

Regarding the best friendships listed by the participants, only 8.2% of them (n = 8) said they had no better or best friends. The majority of the sample said they had one (9.2%, n = 9) or more than one best friend (82.6%, n = 81). Among those who reported having best friends, the mean number of best friends indicated was 3.67 (SD = 1.64), with a minimum of one best friend and a maximum of six (the limit given in the questionnaire).

The other results concerned the best friendship chosen by the participant (in the cases where the individual reported not having better or best friends, they were asked to choose a friendship that stood out more). Table 1 summarizes the characteristics of the best friendships evaluated. The length of the best friendship ranged from a minimum of 13 months to a maximum of 360 months. Regarding the age of the participants, those who lived in the same city as their best friends were significantly younger: they had a mean age of 22.64 years (SD = 3.14), whereas those living in different cities had a mean age of 23.93 years (SD = 2.67), t_96.98 = 1.91, p = 0.060.

In the analysis of the qualities of the best friendship, indices were created based on the indications provided by Souza and Hutz (2007). For each function of friendship investigated (emotional security, intimacy, reliable alliance, companionship, help and self-validation), an index was created with a range from zero to 20, since each factor is composed of five items of the instrument that can receive a value from zero to four each, according to a Likert scale ranging from "never" (0) to "always" (4). The scale of positive feelings and scale of negative feelings related to the friendship were each evaluated as single factors. Thus, two indices were created: the positive feelings index regarding the best friendship (PFFI), with a range of zero to 52, corresponding to the 13 items of the scale, and the negative feelings index (NFFI), with a range of zero to 72.

In the sample, the reliable alliance factor had the highest mean score among the functions of friendship (M = 19.20, SD = 1.58), followed by the intimacy factor (M = 17.41, SD = 2.50). The third factor with the highest mean score was companionship (M = 17.24, SD = 2.33), followed by the emotional security (M = 16.84, SD = 2.91) and self-validation (M = 16.53, SD = 3.14) factors. The factor with the lowest mean in the total sample was the help factor (M = 15.95, SD = 3.03). For the PFFI, the mean of the sample was 50.57 (SD = 2.57), and for the NFFI, 13.78 (SD = 7.38).

Table 2 presents the Pearson correlation analysis regarding age, number of intimate friendships, best friendships, and functions and qualities of the best friendships of the participants, showing the values and significance of these correlations. The number of intimate friendships presented a significant negative correlation with the age of the participants. Similarly, a significant negative correlation was found between the age and the number of intimate friends in the same city. Considering only the best friendships listed, there was also a significant negative correlation with age, although weaker.

The reliable alliance factor demonstrated a significant positive correlation with the length of the best friendship and a significant negative correlation with the frequency of personal contact with the best friend. Help was also significantly positively correlated with the contact with the best friend. Finally, self-validation correlated significantly and positively with the age of the participants (Table 2).

All six functions of friendship showed significant positive correlations with each other. Concerning the feelings in relation to the best friendship, the PFFI was significantly positively correlated with the functions intimacy, companionship, reliable alliance and help. Furthermore, the PFFI also significantly negatively correlated with the NFFI. The NFFI, in turn, correlated significantly only with the function companionship, with negative valence (Table 2).

Discussion

Regarding the intimate friendships listed by the participants, the results showed a significant percentage of friendships with people of the same sex. As previously mentioned, between adults, friendship is characterized by homogeneity in diverse aspects, among them gender (Souza & Hutz, 2008b). The data found in the present study regarding this aspect are similar to the empirical findings of another study with a sample of Brazilian adults (Souza & Hutz, 2008a). Other aspects such as age, the city where they live, the education and the occupation showed the same pattern of homogeneity among adult friends. In the analysis of the networks, groupings of younger people and older people were perceived in the friendship indications of the participants. In relation to the place where they live, the majority of the participants resided in the city of the study (86.9%). Regarding education and occupation, the sample was mainly composed of people with incomplete higher education (75.5%), who reported their occupations as student or trainee (67.3%).

Conversely, some data from the present study were contrary to this similarity filter. Aspects such as the relationship status and the religion did not show homogeneity in the sample. For the analysis of the networks, for example, the participants who were committed indicated non-committed friends as often as they indicated other committed friends, and vice versa. Such results would suggest that friendship between adults actually follows homogeneity, however, perhaps not so comprehensively, with aspects existing that escape this tendency. However, these differences may also have been given because of issues specific to the form of sampling used - RDS. For example, the study by Souza and Hutz (2008a) was based on a convenience sample with a high number of university students participating, finding mixed results in this issue. Further studies are needed to add data on these issues to the literature of the area.

The words most frequently cited by the participants as the first that came to mind when they thought of friendship relationships were in agreement with the studies discussed in the literature review concerning which essential characteristics are constituents of this form of interpersonal relationships (Asher et al., 1996; Bukowski et al., 2009; Duarte & Souza, 2010; Krappmann, 1996; Mendelson & Aboud, 1999). These words can be identified within the three friendship character spheres discussed: social (companionship, trust, sincerity, honesty, respect, loyalty, commitment), instrumental (support, help) and emotional (love, affection, tenderness). Although the data of the present study do not allow generalizations, they can be analyzed as examples that indicate that the representation that young adults have of friendship seem to go together with the theoretical understanding about friendship relationships.

In this study, the younger participants reported more intimate friends than the older participants. This finding was in line with the idea that the number of friends is at its peak in the initial years of adulthood and will reduce with advancing age (DeSousa & Cerqueira-Santos, 2011). The association encountered between a greater number of intimate friends living in the same city for the younger participants suggests that most closer friendships are formed before adulthood. Thus, the hypothesis arises that, for various reasons related to the characteristics of this phase of the lifecycle, people can move out of the city where they live, however, still carry with them their old friendships, which would lead to the older participants indicating fewer friends residing in the same cities as themselves.

The mean length of the best friendship of the participants was over nine years. Considering that the mean age was 23 years, the mean length of the best friendship reached more than a third of the mean age of the participants. It can be seen that, for a relationship to be considered a best friendship, people often consider it to be a lasting relationship. If we compare, for example, the length of the best friendship to the mean of the length of the marital relationships of the participants who were committed (just over two and a half years), it can be noted that, in this sample, the best friendships were relationships of longer duration that the loving relationships. Although the data cannot be generalized for the general population, this result suggests that differences exist regarding the duration for different types of interpersonal relationships.

With respect to the locations where the participants met their best friends, the most indicated contexts were those that provide a high frequency of contact, such as school, family and neighbors. These are fertile environments for the formation of friendships and for their development, which includes providing the passage from close friendship status to that of best friendship (Souza & Hutz, 2008a). Dealing specifically with the school as the place most mentioned, Bukowski et al. (1996) claim that the school years favor the development of lasting friendships due to the emergence of aspects such as loyalty and trust, arising from the social, emotional and cognitive development of the individuals during this phase.

Furthermore, the majority of the participants lived in the same city as their best friend. For Souza and Hutz (2008a), the residential proximity is an important environmental factor for the formation of friendships and also for their development, including the passage from the status of close friendship to best friendship. In the sample studied, the participants residing in different cities to their best friends were significantly older. The hypothesis can once again be proposed that part of the closest friendships is formed before adulthood and sometimes the needs of adulthood lead the individual to change, causing them to reside in a city different from that of their better friend.

Another result found indicated that, in the sample, the longer the participants and their best friends knew each other, the stronger the reliable alliance function was. This result supports the idea that older friendships are strengthened, showing specifically which characteristic is strengthened - in this case, loyalty and availability for the best friend. There was also an inverse correlation between the reliable alliance function and the personal contact with the friend. This finding can be interpreted in the following way: a better friendship with a highest level of alliance between the friends dispenses with the need for very frequent contacts, since both already know how loyal and committed they are to each other. The help function was shown to be directly related to the frequency of contact with the best friend, both personally and by telephone. This is probably because, to request help from a friend, it is necessary to contact them - whilst the commitment of the reliable alliance, after it has been solidified, need not be constantly reiterated.

The fact that all six functions of friendship showed positive correlations with each other and with the general index of friendship functions lends support to the following premise: the higher the level of best friendship in a particular essential function of this type of relationship, the higher the levels will probably be in other functions of this friendship. Good friendships in one particular aspect are usually good friendships overall, and friendships that lack one of the functions also tend to present other fragile aspects. Although they may be separated for the purposes of analysis, all the functions are recurrent in friendship relationships and there is a very strong relationship between them (Duarte & Souza, 2010).

The functions more directly related to positive feelings in relation to the best friendship were the functions of companionship and intimacy, followed by functions of reliable alliance and help. The functions of emotional security and self-validation did not present strong relationships with the positive feelings toward the best friendship. Further studies may try to investigate the causes of these differences between the specific functions of friendship and the positive feelings experienced by people in relation to the friend.

As might be expected, the positive feelings associated with friendship were also inversely related to the negative feelings. Conversely, the result is quite interesting in that it showed that the index of negative feelings was negatively correlated only with the companionship function. According to this finding, the negative feelings toward the friend do not, in general, seem to influence the friendship functions or vice versa. For example, better and stronger friendships can present the same level of negative feelings as the most fragile friendships. This result suggests that it is possible to feel negative emotions regarding the friendships without this affecting how the friendships respond to the majority of their functions.

Final Considerations

The present study aimed to investigate intimate friendship relationships of the social network of young adults. Attention was given to the need indicated by Garcia (2005) for more theoretical and empirical studies regarding friendship relationships in Brazil and to the deficiency discussed by Bukowski et al. (1996) concerning the consideration of the notion of the social network in studies of friendship.

Regarding the configuration of the intimate friendship relationships within the social networks, results were found that were consistent with the idea of the filter of similarity and homogeneity between adult friends (Souza & Hutz, 2008b) in several aspects, mainly in the gender of the close friends; although the data also indicated some differences. This contradiction may be related to the sampling technique used for the recruitment of the participants being different from the techniques used by other studies on the subject with adults (Duarte & Souza, 2010; Souza & Hutz, 2008a). New studies may help clarify these differences.

The present study presents limitations. The main limitation concerns the fact that the sample, although it enabled the analysis of the friendship relationships within a social network, is not representative of the population of young adults in general. Thus, generalizations can not be made without reservations and the conclusions drawn from the discussion of the results should always take into consideration possible peculiarities of the sample. This limitation should be seen as an incentive for future studies in the area. In particular, studies using random sampling techniques and ones that investigate samples representative of the general population would be of great interest to increase the power of generalization of findings in the area and to comprehend the differences between the results of more focused studies.

New work could also focus on other components of human development in the study of friendship relationships, for example, individual characteristics such as the personality of the individuals and their ontogenetic stories. Furthermore, another important aspect that was not covered in depth in this study concerns the role of the internet, cyber relationships and virtual social networks in friendship relationships. Studies show that the use of the internet affects different aspects of the friendships of youth, such as the network of friends, contact, communication, intimacy and closeness (Pylro, 2007). Therefore, studies would be useful which investigate in depth the role of the online world in friendship relationships in adulthood and young adulthood, since the expectations related to friendship vary throughout the lifecycle (Bukowski et al., 2009).

The results presented raise new questions and imply the need for future investigations that analyze in more detail some of the points presented here, as well as covering other aspects not encompassed in the present work. The participants of this study possessed some personal contextual characteristics and the results encountered can be conditioned to this context. Further studies with socially vulnerable populations, non-urban populations, or populations of different social classes, for example, may encounter different results, therefore suggesting that replications address different contexts of human development for the friendship relationships in young adulthood.

References

Neiva-Silva, L. (2010). Estudo comportamental com crianças e adolescentes em situação de rua em Porto Alegre e Rio Grande: Uso da técnica de Respondent Driven Sampling (RDS) para identificação de comportamentos sexuais de risco e uso de drogas - Relatório Final de Pesquisa. Relatório não publicado, Universidade Federal do Rio Grande do Sul, Porto Alegre.

Souza, L. K., & Hutz, C. S. (2007). A qualidade da amizade: Adaptação e validação dos questionários McGill. Aletheia, (25), 82-96.

Received: Jan. 25^th 2011

1^st revision: Mar. 23^rd 2012

2^nd revision: Apr. 20^th 2012

Approved: Aug. 27^th 2012

Diogo Araújo DeSousa is M.Sc. student in Psychology of the Post-graduate Program at the Universidade Federal do Rio Grande do Sul.

Elder Cerqueira-Santos is Associate Professor of the Departamento de Psicologia at the Universidade Federal de Sergipe.

• Abreu-Rodrigues, M., & Seidl, E. M. F. (2008). A importância do apoio social em pacientes coronarianos. Paidéia (Ribeirão Preto), 18(40), 279-288. doi: 10.1590/S0103-863X2008000200006
• Adams, R. E., Santo, J. B., & Bukowski, W. M. (2011). The presence of a best friend buffers the effects of negative experiences. Developmental Psychology, 47(6), 1786-1791. doi: 10.1037/a0025401
• Argyle, M. (2001). The psychology of happiness (2nd ed.). New York: Taylor & Francis.
• Asher, S. R., Parker, J. G., & Walker, D. L. (1996). Distinguishing friendship from acceptance: Implications for intervention and assessment. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The company they keep: Friendships in childhood and adolescence (pp. 366-405). Cambridge, United Kingdom: Cambridge University Press.
• Bukowski, W. M., Newcomb, A. F., & Hartup, W. W. (1996). Friendship and its significance in childhood and adolescence: Introduction and comment. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The Company they keep: Friendships in childhood and adolescence (pp. 1-18). Cambridge, United Kingdom: Cambridge University Press.
• Bukowski, W. M., Motzoi, C., & Meyer, F. (2009). Friendship as process, function, and outcome. In K. H. Rubin, W. M. Bukowski, & B. Laursen (Eds.), Handbook of peer interactions, relationships, and groups (pp. 217-231). New York: Guilford.
• Daudt, P. R., Souza, L. K., & Sperb, T. M. (2007). Amizade e gênero nos conflitos de pré-escolares. Interpersona: An International Journal on Personal Relationships, 1(1), 77-95.
• DeSousa, D. A., & Cerqueira-Santos, E. (2011). Redes sociais e relacionamentos de amizade ao longo do ciclo vital. Psicopedagogia, 28(85), 53-66.
• DeSousa, D. A., & Cerqueira-Santos, E. (2012). Relacionamentos de amizade e coping entre jovens adultos. Psicologia: Teoria e Pesquisa, 28(3), 345-356. doi: 10.1590/S0102-37722012000300010
• Duarte, M. G., & Souza, L. K. (2010). O que importa em uma amizade? A percepção de universitários sobre amizades. Interpersona: An International Journal on Personal Relationships, 4(2), 271-290.
• Fehr, B. (1996). Friendship processes Thousand Oaks, CA: Sage.
• Garcia, A. (2005). Psicologia da amizade na infância: Uma revisão crítica da literatura recente. Interação em Psicologia, 9(2), 285-294.
• Goel, S., & Salganik, M. J. (2009). Respondent-driven sampling as Markov chain Monte Carlo. Statistics in Medicine, 28(17), 2202-2229. doi: 10.1002/sim.3613
• Gomes, L. G. N., & Silva Júnior, N. (2007). Experimentação política da amizade: Alteridade e solidariedade nas classes populares. Psicologia: Teoria e Pesquisa, 23(2), 149-158. doi:10.1590/S0102-37722007000200005
• Krappmann, L. (1996). Amicitia, drubja, shin-yu, philia, freundschaft, friendship: On the cultural diversity of a human relationship. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The company they keep: Friendships in childhood and adolescence (pp. 19-40). Cambridge, United Kingdom: Cambridge University Press.
• Lisboa, C., & Koller, S. H. (2003). Amizade e vitimização: Fatores de risco e proteção no contexto do grupo de iguais. Psico (Porto Alegre), 34(1), 71-94.
• Mendelson, M. J., & Aboud, F. E. (1999). Measuring friendship quality in late adolescents and young adults: McGill Friendship Questionnaires. Canadian Journal of Behavioural Science, 31(2), 130-132. doi:10.1037/h0087080
• Oliveira, M. C. S. L., Camilo, A. A., & Assunção, C. V. (2003). Tribos urbanas como contexto de desenvolvimento de adolescentes: Relação com pares e negociação de diferenças. Temas em Psicologia, 11(1), 61-75.
• Pereira, F. N., & Garcia, A. (2007). Amizade e escolha profissional: Influência ou cooperação? Revista Brasileira de Orientação Profissional, 8(1), 71-86.
• Peron, S. I., Guimarães, L. S., & Souza, L. K. (2010). Amizade na adolescência e a entrada na universidade. Estudos e Pesquisas em Psicologia, 10(3), 664-681.
• Pylro, R. (2007). Amizade.com: Amizade e adolescência na era da Internet. Dissertação de mestrado não publicada, Universidade Federal do Espírito Santo, Vitória.
• Rawlins, W. K. (1992). Young adult friendships. In Friendship matters: Communication, dialectics, and the life course (pp. 103-123). New Brunswick, NJ: Transaction.
• Santana, J. J. R. A., Zanin, C. R., & Maniglia, J. V. (2008). Pacientes com câncer: Enfrentamento, rede social e apoio social. Paidéia (Ribeirão Preto), 18(40), 371-384. doi:10.1590/S0103-863X2008000200013
• Souza, L. K., & Hutz, C. S. (2008a). Amizade na adultez: Fatores individuais, ambientais, situacionais e diádicos. Interação em Psicologia, 12(1), 77-85.
• Souza, L. K., & Hutz, C. S. (2008b). Relacionamentos pessoais e sociais: Amizade em adultos. Psicologia em Estudo, 13(2), 257-265. doi:10.1590/S1413-73722008000200008

Publication Dates

• Publication in this collection
  15 Mar 2013
• Date of issue
  Dec 2012

History

• Received
  25 Jan 2011
• Accepted
  27 Aug 2012
• Reviewed
  20 Apr 2012