Who is Friends with Whom?
The degree to which people from different social strands have relations with each other indicates social cohesion. Therefore, the question of who is friends with whom is nontrivial. Compared to marriage, friendship is a non-institutionalized relationship: there is no formal start of a friendship and one can break off or change a friendship without notifying any third parties.
A generally accepted principle is that people prefer being friends with others who are like themselves. Interactions with similar others are rewarding (Homans 1984: 158). Seminal research has been done by Lazarsfeld and Merton (1954), who discuss value and status homophily affecting the selection of friends. Yet explanatory mechanisms differ among theories: demand side theories focus on individual preferences, while supply-side theories are directed to the distribution of meeting chances in society. Important demand side perspectives are balance theory (Heider 1946) and the theory of social capital (on a micro level, Flap 2004). Important supply side perspectives are (macro ) structuralism (Blau 1977) and focus theory (Feld 1981). An integration of both perspectives is the choice constraint approach (Fischer et al. 1977).
With regard to demand side perspectives, balance theory states that a friendship between two actors depends on their relationship with a third party. If two actors have a positive relation to a third party, they are likely to also form a positive relationship with each other. Yet if one actor has a positive and the other a negative relationship to a third party, the positive relationship is less likely. The theory assumes that imbalance in relationships produces a strain, which people reduce by changing or breaking off relationships. Importantly, balance theory takes existing friendships into account when it comes to the decision to create a new one. Further, social capital theory states that people become friends if they face a common future, if they are in one or another way dependent on convivium each other, or if they have invested in the relationship formerly.
The most prominent supply side theory is Blau’s (1977) structural approach, which abandons a pure micro level exchange perspective and takes macro structures into account. Patterns of homophily are dependent on relative group sizes in the population. Interestingly, if different individual attributes are not closely correlated, intergroup associations can result despite the preference for in group associations. Related to the importance of numerical distribution is the argument on geo graphical proximity. Proximity reduces costs of interaction and facilitates the emergence of mutual trust. Furthermore, focus theory generalizes from numbers and places and assumes that individuals who share foci of activity have higher chances for meeting, and therefore greater chances for becoming friends (Feld 1981).
Lastly, the choice constraint approach emphasizes that relationships are the result of individual choices made under social constraints (Fischer et al. 1977). People choose to construct and maintain social exchanges with some of those whom they encounter and they make this choice on the basis of weighing rewards and costs.
Like marriage, friendship is a relationship that occurs among those who are similar in relevant social dimensions, like age, education, class, ethnicity, and religion (Laumann 1973; Fischer et al. 1977). Yet, unlike marriages, cross sex friendships are a rare phenomenon.
With regard to age, Fischer et al. (1977) found that 38 percent of respondents’ close friends were within 2 years of their age. Friendships are class sensitive in general, yet similarity is highest within higher classes. With regard to ethnicity, friends are also remarkably similar in ethnicity. Esser (1989) found that even second generation immigrants in Germany had friend ship networks that were largely in their own group. Cohen (1977) showed that Jewish and black people have the highest tendency for in group association and Scottish people the lowest. Further, Protestants are somewhat more in group oriented than Catholics. Little research has been done on the question of whether similarity in one social dimension is associated with similarity in another. An exception is the study by Jackson (1977), which shows that friends who work in the same economic sector also have a higher chance to be similar in education, age, and ethnicity.
A number of studies focus on the question ‘‘who has friends’’ rather than on who is a friend to whom. Interesting findings have been provided, showing that structural characteristics strongly influence people’s associations. For example, married people have fewer friends than unmarried people and higher educated and richer people have more friends. With increasing age, the number of friends first increases, but then decreases. Lastly, numbers of friends also differ between people from different countries.
Tests of balance theory largely corroborate the ‘‘friends of friends are friends’’ proposition. Yet the difference between cognitive balance (indicating a state of mind) versus structural balance (indicating the structure of personal networks) should be taken more seriously in research.
Research into friendship and social capital mainly studies what friends can do for each other in order to achieve important individual goals. It has been shown that strong ties are not important in attaining things like a job (Granovetter 1995). Weaker ties are more important for these kinds of achievements. Fischer (1982) showed, in addition, that friends are not important for monetary transactions or any other material exchanges. Furthermore, while friendships are important for all kinds of social activities, they are not that important for matters of serious advice.
Blau’s macro structural theory has been tested by Blum (1985) for socializing relationships and with regard to ethnic and religious heterogeneity. Blum demonstrated that while there are preferences for ingroup association, structural conditions exert substantial constraints. Heterogeneous populations promote intergroup relationships. McPherson and Smith Lovin (1987) also provide a test of Blau’s theory and find evidence for the higher importance of group composition compared to individual preferences.
Concerning the social settings from which friends are drawn, Feld (1981) found that 68 percent of the relationships of the respondents in his study were formed in a shared setting with roughly a third in work or voluntary associations (see also Marsden 1990). Social settings differ in the degree to which they enhance friendship formation. Friends drawn from childhood are most similar in age, friends drawn from the work setting are most similar in occupational level, and friends drawn from a kin setting are above all similar in ethnicity ( Jackson 1977). Furthermore, the importance of settings for recruiting friends differs between classes and life stages and also between countries.
Friends are remarkably similar to each other in various dimensions, and the tendency to associate with similar others differs according to age, education, class, ethnicity, and religion. The degree to which similarity in one dimension is associated with similarity in another is rarely investigated. Furthermore, friendships are drawn from different sources and the dimension on which friends are similar partially depends on the source from which the friends are recruited.
Both perspectives, demand as well as supply side, have been corroborated in research, and although there is some evidence that the supply side might be even more important, it is not clear what the relative importance of preferences and constraints in friendship choice would be. It is furthermore noteworthy that systematic empirical accounts on ‘‘convivium’’ are somewhat dated and restricted to the US. Future research has the task to overcome these shortcomings. In addition, the assumption that everybody needs and has friends might not be true. Lastly, most research concentrates on friendship dyads and not on networks, thereby disregarding the fact that friendship relationships are not exclusive relationships, but are embedded in social networks.
- Blau, P. (1977) Inequality and Heterogeneity. Free Press, New York.
- Blum, T. (1985) Structural Constraints on Interpersonal Relations: A Test of Blau’s Macrosociological Theory. American Journal of Sociology 91(3): 511 21.
- Cohen, S. M. (1977) Socioeconomic Determinants of Intraethnic Marriage and Friendship. Social Forces 55(4): 997 1010.
- Esser, H. (1989) The Integration of the Second Generation: Toward an Explanation of Cultural Differences. Zeitschrift fu¨r Soziologie 18(6): 426 43.
- Feld, S. L. (1981) The Focused Organization of Social Ties. American Journal of Sociology 86: 1015 35.
- Fischer, C. S. (1982) To Dwell Among Friends. University of Chicago Press, Chicago.
- Fischer, C. S. et al. (Eds.) (1977) Networks and Places: Social Relations in the Urban Setting. Free Press, New York.
- Flap, H. (2004) Creation and Returns of Social Capital: A New Research Program. In: Flap, H. & Volker, B. (Eds.), Creation and Returns of Social Capital: A New Research Program. Routledge, London, pp. 3 23.
- Granovetter, M. (1995 ) Getting a Job. Cambridge University Press, Cambridge.
- Hallinan, M. (1990) The Formation of Intransitive Friendships. Social Forces 69(2): 505 19.
- Heider, F. (1946) Attitudes and Cognitive Organization. Journal of Psychology 21: 107 12.
- Homans, G. C. (1984) The Human Group. Harcourt, New York.
- Jackson, R. M. (1977) Social Structure and Process in Friendship Choice. In: Fischer, C. S. et al. (Eds.), Networks and Places: Social Relations in the Urban Setting. Free Press, New York, pp. 59 78.
- Laumann, E. O. (1973) Bonds of Pluralism: The Form and Substance of Urban Social Networks. Wiley Series in Urban Research. Wiley & Sons, New York.
- Lazarsfeld, P. & Merton, R. K. (1954) Friendship as a Social Process: A Substantive and Methodological Analysis. In: Berger, M., Abel, T., & Page, C. H. (Eds.), Freedom and Control in Modern Society. D. Van Nostrand, Toronto, pp. 18 66.
- McPherson, M. & Smith Lovin, L. (1987) Homophily in Voluntary Organizations: Status Distance and the Composition of Face-To Face Groups. American Sociological Review 52(3): 370 9.
- Marsden, P. (1990) Network Diversity, Substructures and Opportunities for Contacts. In: Calhoun, C., Meyer, M. W., & Scott, W. R. (Eds.), Structures of Power and Constraints. Cambridge University Press, Cambridge, pp. 397 410.