Aspects of gender in social networks

Document Sample
Aspects of gender in social networks Powered By Docstoc
					In Czagany, L. – Garai, L. 2004: Social identity, information and markets.
Proceeding of Faculty of Economics and Business Administration 2004. JATEPress,
Szeged (Under publication)

                 Aspects of gender in social networks
                                      Zsófia Kürtösi1

Key words: network perspective, gender differences, intraorganizational networks

1. Introduction

       Analysing relationships, i.e. the network perspective can be considered as one
of the youngest coherent theoretical-methodological trends of sociology. Other
social sciences have also discovered its novel methods and approaches and have
been using them productively in their own fields of research to analyse the
relationships between different social actors and groups, to examine their structure
or to explain their behaviour.
       In a broad sense, a social network consists of a finite set of actors and the
relation or relations between them (Wasserman–Faust 1994). These social actors
may be either individuals, or organisations, or any other collective social units, and
the collection of ties of a specific kind among members of a group is defined as a
relation. These ties can have different contents, thus they can involve a material or
immaterial resource transfer (eg loan, advising, information, exchange of goods), a
biological relationship (kinship, marriage), a physical link, an association, adherence
to a group, a value judgement (friendship, expression of respect) or even a formal
       This paper solely concentrates on the forms of interaction between individuals
and aims to present its possible gender differences, and networks in work
organizations in particular. I touch upon those network differences from previous
stages of the lifecourse which might affect the different behaviour of the sexes and
the way how they establish contacts at a workplace.

2. Network perspective

      One of the fundamental features of the network perspective – distinguishing it
from other more coherent theoretical trends – is that instead of individual actors and

 Zsófia Kürtösi, assistant professor, University of Szeged, Faculty of Economics and Business
Administration (Szeged)
their characteristic attributes it concentrates on the relationships between these
actors. A network approach suggests that social processes and individual results are
basically defined by the network patterns between the actors (Blau 1989,
Granovetter 1996) even with so strongly individual-related characteristics as gender.
A network prospective stresses social embeddedness hence distinguishing between
gender conceived as a result of network interactions and biologically determined
       Unlike ego-centred explanations, a network approach explains gender
differences in the responses given to different social situations with the differences
in networks and with the regularities in their functioning. According to theory, the
changes in people’s sets of relationships surrounding them in each stage of their
lives induce a change in their seemingly stable individual characteristics, attitudes
and values. A possible stability, on the other hand, can be explained by the same
stability of the fundamental features of social networks (Smith-Lovin–McPherson
1993). At the same time, a certain feedback can be identified as well, since it is not
only relationships that affect an individual’s possibilities and, consequently, the later
stages of his or her lifecourses but life situations in turn have a feedback on an
individual’s social network.
       Research into networks taking gender as a basic category of analysis into
consideration do not have a very long history, as such investigations were launched
in the 70s and have focused primarily on examining differences and the extent of
these differences in social networks of men and women, exploring the possible
reasons for such differences including those possible influences that social networks
can exert on an individual’s lifestages and opportunities. Besides investigations into
ego networks, so-called total network analyses have also become widespread, aimed
at exploring all the relationship systems within a well-defined group of actors.
Organizational research can serve as a good area for such investigations because
membership in an organization clearly defines network boundaries. A gender-based
analysis of intraorganizational networks among others tries to shed light on the
network strategies of men and women, exploring whether their behaviour in an
organization is different and if yes, to what extent their results can be regarded as
consequences of network patterns, whereas an investigation of interaction patterns
on different organizational levels extends the analysis to the behaviour of minority-
majority groups within a network as well.
       Below I describe some interaction characteristics that evolve as early as
childhood and may determine how relationships change in men and women’s
subsequent lifecycles.
3. Gender differences of social networks

3.1.   Childhood networks
        If we assume that through an access to information and opportunities, the
network characteristics of women and men as well as their different position in
networks influence the later stages of their lifecourse, then we have to take a closer
look at those differences between the two sexes that appear at an early stage of their
lives and which differences determine later network patterns and their networking
        In the investigations carried out into childhood interactions, researchers first
of all took notice of the homophilic nature of relationships: children tend to play
together with others who are similar to them. Similarity in this case means similar in
terms of sex; it is interesting to note, however, that although psychologist estimate
that awareness of sexual identity appears at the age of 4 or 5 (Cole–Cole 2001),
preference for same-sex friendships has already been proved at a younger age (La
Freniere et al 1984). This leads us to think that gender preference probably stems
from behavioural compatibility, but due to the small number of research studies we
cannot reject those explanatory theories either which are based on the capacity to
identify sex (a child’s own and another child’s sex) according to which children
prefer “like-me” playmates; nor can we dismiss those assumptions that focus on the
role of direct reinforcement, the social approval from adults for playing with same-
sex peers (La Freniere et al 1984, p. 1963.).
        Further research was carried out into differences arising from the size of
playgroups: analysing the complexity of games played by 10-11-year-old children
Lever (1978) found that the size of girl and boy groups at a young age is different:
whilst boys seem to prefer bigger playgroups, the size of girl groups in the
investigations seldom reached 10 members. Girls claimed to prefer the company of a
single friend rather than that of a smaller group. The findings of Benenson’s (1993)
research involving 4- and 5-year-old children confirmed girls’ preference for dyadic
relationships as well. Lever (1978), however, also notes that girls who are more
likely to participate in complex games requiring large groups tend to prefer larger
friendship groups. Since childhood games, and team games in particular, can play a
relevant role in promoting strategic thinking, practicing co-operative and
competitive roles as well as in acquiring such social skills that can prove to be useful
in organizational bureaucracies characterized by fierce competition and hierarchical
conditions. These gender differences – according to Lever – can affect their success
later in their careers. Among the causes of sexual differences that are perceptible in
games Lever (1978) identified cultural and historical reasons, above all.
        In addition to research focusing on gender differences of different group
sizes, there have also been investigations as to how children at this early stage of
their lives react to so-called intransitive emotional relationships. Emotionally
unbalanced triads eg when A is fond of B, B is fond of C but A does not like C,
induce emotional tension or anxiety urging an individual to act in order to restore the
emotional balance (Heider 1946, Hallinan–Hutchins 1980). In the effort to restore
the emotional balance, sex differences can be noticed: while boys are more likely to
resolve intransitivity by making new friendships, girls are more likely to delete their
intransitive relationships (Eder–Hallinan 1978) and choose closed dyadic friendships
more frequently. Also, the sex of the third person wishing to join this same-sex dyad
is of determining importance, since children tend to tolerate cross-sex intransitive
relationships better than a same-sex intransitive contact. But besides this, they also
tend to resolve intransitivity by adding a same-sex relationship (Smith-Lovin–
McPherson 1993, p. 231.).
        The above findings call our attention to the fact that network patterns of an
early stage of a lifecourse may affect the social skills and relationship preferences as
an adult and, to some extent, can be held responsible for a separation of women and
men’s social world.

3.2.    Adult networks
       In most cases adult networks preserve those characteristics which are
perceptible in childhood, but different stages of their lives, various situations but
especially parenthood and a disproportionate distribution of paid and domestic work
generally further increase differences in men and women’s networks (Wellman
       Homophily i.e. the choice of like-me friendships is typical of not just
childhood but of adult networks as well: members of a network preferably establish
contacts with those who have similar characteristics in terms of age, sex, race,
nationality or educational background. But when examining gender-based
homogeneity of ego-centred adult networks, we have to consider the rate of family
and non-family as well as neighbourhood relations, for these increase the age- and
gender-based heterogeneity of an individual’s relationships (Marsden 1991).
According to US investigations, we find more family and neighbourhood ties in
women’s discussion networks, whereas in men’s network the number of work
contacts is higher thus making this network more homogeneous2 (Wellman 1985,

  This gender difference can obviously originate from the peculiarities of the US labour market as well,
since certain forms of employment encouraging flexibility of the labour market such as part-time
employment (mostly among women) are much more widespread in the United States than eg in
Hungary. This assumption appears to be supported by a comparative study carried out in 1986 on a
representative sample involving seven countries (including Hungary) which showed a considerable
overlapping between friendship and work ties being a typically Hungarian phenomenon (Utasi 1991, p.
182.). Researchers explained this by the unusually long working hours, a common practice in Hungary
for both sexes at that time, since a considerable part of those who worked in the second economy had
no time to maintain friendships outside the workplace (Utasi 1991).
Smith-Lovin–McPherson 1993). A Hungarian survey by TÁRKI3 in 1999 conducted
with a view to explore intimate relationships confirmed that gender-based
homogeneity is less typical of family relationships, while non-family relationships
turned out to be rather homogeneous in terms of sex (in the entire sample the rate of
same-sex respondents among non-family relationships exceeded 76 per cent)
(Albert–Dávid 1999). Naturally network contents (eg friendship, communication,
social support) may affect gender-based homogeneity of networks while with most
network contents, gender-based homophily is a general trend.
        A part of the gender-based differences in network sizes, which are
characteristic of one’s childhood, are preserved in adult networks as well. Although
in the size of women’s and men’s networks no difference has been found (Smith-
Lovin–McPherson 1993), a preference for dyadic relationships is stronger with
women in adulthood as well. A research by Stokes and Levin (1986) revealed that
while for men the feeling of loneliness is mostly judged by the density of their social
networks (the denser a network, the less they feel lonely), women judge this feeling
of solitude based on the lack or presence of more intimate, dyadic relationships.
        A significant part of gender differences in adult network characteristics stem
from an uneven distribution of productive and reproductive tasks. Although in
advanced societies women get increasingly involved in paid work (except for ex-
socialist countries where a slight decline can be seen) and their position in the labour
market is strengthened, their role in the family and their sole responsibility for
looking after the old ones, the ill and children is usually not questioned. This
division of labour obviously influences their networks, since a greater participation
in child rearing introduces women to social circles that, above all, facilitate the flow
of information, values and opportunities linked to family and relatives, whilst men’s
social networks provide information about and access to career, job and free-time
activities (Smith-Lovin–McPherson 1993). This seems to be supported by the fact
that in the results of Wellman’s (1985) investigations single women and men’s
social networks do not differ radically, thus it suggests that sex differences of
networks disappear or can at least decrease substantially if we control for marital
status, the lack or presence of children and employment status. Different
circumstances affect the changes in relationships, which by facilitating access to
information and other resources again affect life situations (eg to possible career
        People take their differences in network structures and their inclination to
interactivity into those work organizations where they spend a considerable part of
their lives, but the geographical or dimensional limitations of these organizations
reduce the possibilities of the members for developing and implementing their
network strategy.

    TÁRKI Social Research Center Inc.
4. Networks in work organizations

4.1.   Features of workplace networks
         Like extra-organizational networks, workplace networks are rather varied
and can be manifold in terms of their contents. As a rule they comprise
communication, advising, career-supportive and friendship networks, or respect
networks, that is they include both instrumental and expressive elements. A network
is called instrumental if “actors contact one another in efforts to secure valuable
goods, services, or information” (Knoke–Kuklinski 1982, p. 16.). Instrumental ties
are in general weaker, yet very important sources of technical information that
support individuals in doing their work more effectively. In expressive networks
individuals express their feelings of attraction, respect or hate to others (Knoke–
Kuklinski 1982). These networks can primarily secure “emotional” resources and
are stronger and more intimate than instrumental ties. Researchers try to examine
these network contents separately but, owing to organizational limits, overlapping is
frequent; thus intraorganizational friendship networks “behave” differently from
those friendship networks that are not restricted by organizational settings.
Organizational “friendships” are also systems of decision-making: they mobilize
resources, facilitate the flow of information; therefore they are similar in structure to
instrumental networks (Lincoln – Miller 1979).
       Another relevant feature of intraorganizational networks is that these
relationships are largely dependent on not only a person’s individual preferences but
on opportunities and restrictions of the organizational structure as well: accessibility
of a person within the network greatly reduces the options of the individual.
Therefore, intraorganizational networks may differ from extra-organizational ones:
in the latter, and especially with strong ties, inclination to homophily is frequent; in
work organizations, on the other hand, their composition based on individual
characteristics is determined from the outset, which may limit the choices. To test
this, McPerson and Smith-Lovin (1987) examined the similarity of friendship-
relations in voluntary organizations considering four dimensions (sex, age, job,
educational background) and then tried to separate induced homophily deriving from
accessibility restrictions of actors from choice homophily which is a result of
individual preferences. They were looking for the answers to the question to what
extent a perceived homophily being particularly strong from the point of view of sex
and age may be due to the composition of the organization. According to the results
of this research, in terms of sex, age and job, induced homophily was greater than
choice homophily, whereas in the case of educational background the opposite was
true (McPerson–Smith-Lovin 1987, p. 375.). The investigation has also
demonstrated that homophily grows as the size of the group increases, since a bigger
group provides more possibilities for reaching people with similar characteristics.
When examining the like-me principle, in addition to sociodemographic variables,
such special features should also be examined as the time of joining an organization.
Examining communication networks in an organization, Wagner, Pfeffer and
O’Reilly (1984) found that people joining an organization at roughly the same time
form a more coherent group. This is because those who joined earlier have a well-
established communication network, the new members, however, would like to
make contact, so they try to communicate with those, first of all, whose
communication capacities are still free.
        But in work organizations, it is not necessarily a social similarity that
individuals mostly strive towards in their relationships. The advantage for certain
individuals is to maintain contacts with people having higher status, power or
influence, so that they could obtain valuable resources in this manner. A superior
status, power and influence may correlate with some important individual
characteristics (eg sex, race), hence for the members of a particular group homophily
in one dimension might automatically mean homophily in another dimension as well
(Blau 1977); a similarity in personal characteristics facilitates the development of
trust, but it facilitates communication and a reconciliation of individual interests
(Lincoln–Miller 1979).

4.2.   The application of the like-me principle, social identity in a work setting

       Establishing relationships based on similarity is one of the most typical
networking strategies both inside and outside the boundaries of an organization.
How an individual determines the characteristics of similarity largely depends on his
or her immediate surroundings, as the organizational environment is rich in those
social categories in which an individual can locate him or herself.
       In self-categorization those visible categories can play a relevant role in which
an individual’s membership is for life, such as eg colour, visible physical handicap
or gender. Since an individual can be member of several groups at the same time,
the degree of activation of group membership and the dominance of one or the other
group membership develop as a result of several social and cognitive factors
(Smith–Mackie 2001). Generally we can say that within a given organizational
context, the rareness of a category determines its use, hence it is typical for instance
that the gender composition of a particular group affects the use of gender as a
category: group members, who – according to the given subcategory – are
underrepresented, mention their sex as a category of identification more frequently
(Smith–Mackie 2001). The emphasis is primarily on proportions characteristic of the
immediate surroundings rather than on the broader social environment. Changes in
the local conditions can bring about a shift in emphasis from one membership
towards another one, thus a definition of the “self” and “others” should be handled
as a flexible, comparative and relational category (Ashforth–Mael 1989, p. 21.).
According to Mehra et al (1998) if an individual belongs to groups that are
underrepresented from a number of aspects, eg being an Afro-American woman she
is a member of a majority white group mainly made up of men, then the basis for
identification (i.e. whether she identifies herself rather as a woman or as an Afro-
American) is determined by the rareness of the given social categories. Self-
definition and identification with a given group, however, do not only depend on
proportions. Ashfort and Mael (1989) describe organizational situations in which
identification with a group becomes stronger, including for example group prestige
or even the presence of out-groups i.e. of people who do not belong to the group. If
the presence of out-groups is perceived – irrespective of their proportion – it usually
consolidates group boundaries and encourages the assumption of in-group
homogeneity (Ashforth–Mael 1989, p. 25.). Thus the presence of women, for
example, in an area predominantly represented by men can encourage men’s
behaviour patterns that stress gender differences (Kanter 1977). Another factor
promoting group identification is if there is a competitive environment among the
groups (eg within an organization for “scarce” higher positions), because these
situations sharpen boundaries and highlight differences among groups (Smith–
Mackie 2001).
       If an individual sees him or herself as a member of a particular group, this
might influence his or her way of thinking or behaviour towards other group
members or outsiders. A typical consequence of group membership is in-group
favouritism and a discrimination against out-groups (Smith–Mackie 2001). Research
has proved that group members usually remember pieces of information
emphasising intergroup differences better than those highlighting similarities
(Ashforth–Mael 1989, p. 31.). Group membership, i.e. how an individual defines
and assigns him or herself to a particular category, greatly affects the development
of his or her relationships, but most of all, of expressive contacts as well (Mehra et
al 1998). According to Mehra et al (1998, p. 447.) people’s identity and choices of
friendship are based on how they perceive similarities by virtue of distinctiveness.
       As an organization provides individuals with a number of opportunities for
choosing group membership and as the conflicts among these groups originate from
the existence of the groups themselves (Ashforth–Mael 1989, p. 31.), organizations
have to make serious efforts to preserve demographic diversity and integration

4.3.   Gender differences in organizational networks
        There have been several investigations regarding to what extent gender
(besides choices based on the like-me principle) influences relationship networks
within an organization. According to certain widely held views, men can much
better exploit informal interactions for the sake of becoming promoted within an
organization, whereas women attach less importance to informal relationships and
they tend to rely on formal structures (Hennig–Jardim 1977). Others say that it is not
only about women mobilising their informal relationships less to reach their
organizational goals, but also about the fact that men, being a majority group in the
management of business organizations, exclude women from these networks so that
they could maintain their dominance (Albrecht 1983). In Kanter’s (1977) opinion,
however, the separation of men and women’s interaction networks, as proved by
Brass (1985) as well, simply stems from a considerable uncertainty of the
organizational environment resulting in the fact that organizational actors seek to
establish contacts with similar others in order to be able to solve tasks more
effectively: so men are more likely to keep informal relationships with men, and
women do so, in the first place, with women. But Ibarra’s (1992) research revealed
that men and women establish different organizational contacts. While men usually
choose men for both their expressive and instrumental ties, women prefer men for
their instrumental network ties (advising and influence) in order to maximize both
emotional and instrumental resources; for their expressive ties (friendship),
however, they choose women but in their communication network men and women
are represented in equal proportions. Ibarra (1992) also notes that in men’s networks
a greater degree of gender-based homophily does not imply higher multiplexity, i.e.
men choose different male colleagues depending on the different network contents.
       Examining gender differences in networks they also investigated whether it is
women or men who more frequently hold a central position in networks, i.e. who are
the most eminent and important people and whether there is a difference between
the sexes in this respect. In this case, importance and centrality can be interpreted in
a number of ways: first of all, those people can be regarded as important actors of a
network who show the highest degree of network activity, since they are joined by
many other people, or they have a number of ties to others. We must distinguish
between these two types of relationship because choices of relationship are very
often asymmetric: many people call others their friends or respect them for their
professional competence, but they do not necessarily return these choices. In such
cases it might be worth examining how many choose the “choosers” i.e. if person A
is chosen as a friend by a person B who is chosen by many people, then we can
consider person A more prestigious and important than in the case when a less
popular B chooses A (Wasserman–Faust 1994). Those as well can be regarded as
occupying a good position who are able to interact with others very quickly, so
when gathering information they do not have to rely on distant relationships
(accessible only through others), thereby risking a distortion of information.
Cutpoints of the network can also be regarded as holding a good position that is
those who, if withdrawn from the network, would cause it to break up into more
components than it comprised originally. Various centrality and prestige indices can
be calculated for different concepts and it is rather common that researchers
characterize the position of network actors with several indices.
       According to a possible view, those who have a higher educational level, have
more experience and organizational knowledge as well as more expertise, in short,
who have more human capital are the most desirable network targets i.e. occupy a
central position in a network. Since women occasionally accumulate less human
capital, they are less desirable from the point of view of network relationships, thus
they are more likely to occupy a marginal position. Brass’s (1985) investigations,
however, partly seem to contradict the above, demonstrating that women equally
hold central positions in interaction networks just like men (especially in the case
when their proportion in an organization is equivalent to that of men) but not in
„appropriate” networks, i.e. not in interactions involving the dominant coalition4,
thus they are less looked upon as influential persons5.

4.4.    Effects of composition in organizations

       The gender composition of work organizations or their subunits and
organizational levels has been the subject of several investigations into intergroup
social interactions. Kanter’s (1977) “token” situation, which is a consequence of
underrepresentation in terms of numbers, leads to a considerable “visibility” of the
minority group, a critical observation of the majority group and to a decrease in the
number of intergroup interactions. But further research has emphasised that the
operating mechanisms of a “token” situation are influenced by the status of the
groups as well, therefore a minority group enjoying a higher status (eg men) is not
necessarily confronted with a hostile behaviour of a majority group occupying a
lower status, or with an exclusion from all communication chains (Fairhurst–
Snavely 1983). Wharton and Baron (1987) came to a similar conclusion when
examining the effects of gender segregation on the mental “well-being” of men in
work settings with various gender compositions. Their research justified that men
tend to be more satisfied with their work and claim to have more self-confidence in
a work setting dominated either by men or women (their satisfaction appeared to be

  A group, the members of which were high up in the organizational hierarchy and who are
acknowledged by the others as influential persons (Brass 1985).
  Besides sex, obviously innumerable demographic or other factors can influence network position eg
age, the time spent within an organization and even personal characteristics. An investigation by
Mehra, Kilduff and Brass (2001) showed that the type of personality may affect the network position of
individuals. Their investigations focused on people with an open personality who continuously monitor
and strategically manipulate their ego in their relationships with others. The results of their
investigations proved that people with such personal characteristics are more likely to occupy a central
position in social networks and, moreover, their position tends to become increasingly central with time
(Mehra et al 2001, p. 140.).
slightly greater in units dominated by women) but a gender-balanced working
environment leads to a decrease in the general satisfaction with their job (Wharton–
Baron 1987). The findings suggest that a well-balanced gender composition can
limit interactions between men and women (Wharton–Baron 1987, p. 585.), hence
strategies striving to reach an equal proportion of genders in workgroups could even
be harmful, unless they entail interventions aimed at raising the quality and number
of contacts among segregated gender groups. Brass (1985) came to the same
conclusion pointing out that organizations need to endeavour to form integrated
work groups linking isolated male and female interaction networks because this
ensures greater productivity and efficiency for both individuals and organizations.

5. Conclusion

       A network perspective makes it possible to view and treat individual
decisions, actions and results from a broader perspective embedded in the system of
social relations. Sex differences of network formations that can be detected at early
stages of life can affect adult networks of individuals, and by providing possibilities
and placing restrictions they can contribute (besides many other factors) to the
development of different life situations, which in turn affect an individual’s
networks in an endless spiral. On joining an organization, people do not get rid of
their network experience, preferences, or skills from the “outside world”, hence their
behaviour within this organization is determined by these as well. A network
analysis approach and methodology can add valuable elements to an investigation of
the internal and functional processes of an organization. Network researchers
believe that this approach is capable of putting the long-standing problems in a new
light and, in this way, it can broaden the scope of analysis of intraorganizational
gender differences.


Albert     F. – Dávid B. 1999: A bizalmas kapcsolatokról. Lelőhely: Letöltés ideje: 2003.
         május 21.
Albrecht, S. L. 1983: Informal Interaction Patterns of Professional Women. In
      Gordon, J.R. (ed.): A Diagnostic Approach to Organizational Behavior. Allyn
      and Bacon, Boston
Ashforth, B. E. – Mael, F. 1989: Social Identity Theory and the Organization.
     Academy of Management Review, 14, 1, pp. 20-39.
Benenson, J. F. 1993: Greater Preference Among Females Than Males for Dyadic
     Interaction in Early Childhood. Child Development, 64, pp. 544-555.
Blau, P. M. 1977: Inequality and Heterogeneity: A Primitive Theory of Social
      Structure. Free Press, New York
Blau, P. M. 1989: Structures of Social Positions and Structures of Social Relations.
      In Turner, J. (ed.): Theory Building in Sociology. Sage, London
Brass, D. J. 1985: Men’s and Women’s Networks: A Study of Interaction Patterns
      and Influence in an Organization. Academy of Management Journal, 28, 2, pp.
Cole M. – Cole, S. R. 2001: Fejlődéslélektan. Osiris, Budapest
Eder, D. – Hallinan, M.T. 1978: Sex Differences in Children’s Friendship. American
      Sociological Review, 43, April, pp. 237-250.
Fairhurst, G. T. – Snavely, B.K. 1983: A Test of the Social Isolation of Male
      Tokens. Academy of Management Journal, 26, 2, pp. 353-361.
Granovetter, M. 1996: A gazdasági intézmények társadalmi megformálása: a
      beágyazottság problémája. In Lengyel Gy. – Szántó Z. (ed.): A gazdasági élet
      szociológiája. Aula, Budapest
Hallinan, M.T. – Hutchins, E.E. 1980: Structural Effects on Dyadic Change. Social
      Forces, 59, 1, pp. 225-245.
Heider, F. 1946: Attitudes and Cognitive Organization, Journal of Psychology, 21,
      pp. 107-112.
Hennig, M. – A. Jardim 1977: The Managerial Woman. Anchor Press/Doubleday,
      New York
Ibarra, H. 1992: Homophily and Differential Returns: Sex Differences in Network
       Structure and Access in an Advertising Firm. Administrative Science
       Quarterly, 37, September, pp. 422-447.
Kanter, R. M. 1977: Men and Women of the Corporation. Basic Books, New York
Knoke, D. – Kuklinski, J.H. 1982: Network Analysis. Sage, Bevery Hills
La Freniere, P. – Strayer, F.F.– Gauthier, R. 1984: The Emergence of Same-Sex
     Affiliative      Preferences      among        Preschool    Peers:    A
     Developmental/Ethnological Perspective. Child Development, 55, pp. 1958-
Lever, J. 1978: Sex Differences in the Complexity of Children’s Play and Games.
      American Sociological Review, 43, August, pp. 471-483.
Lincoln, J. R. – Miller, J. 1979: Work and Friendship Ties in Organizations: A
      Comparative Analysis of Relational Networks. Administrative Science
      Quarterly, 24, June, pp. 181-199.
Marsden, P. V. 1991: Az amerikaiak fontos beszélgetési hálózatai. In Angelusz R. –
     Tardos R. (ed.): Társadalmak rejtett hálózata. MKI, Budapest
McPerson, J. M. – Smith-Lovin, L. 1987: Homophily in Voluntary Organizations:
     Status Distance and the Composition of Face-to-Face Groups. American
     Sociological Review, 52, June, pp. 370-379.
Mehra, A. – Kilduff, M. – Brass, D.J. 1998: At the Margins: A Distinctiveness
      Approach to the Social Identity and Social Networks of Underrepresented
      Groups. Academy of Management Journal, 41, 4, pp. 441-452.
Mehra, A. – Kilduff, M. – Brass, D.J. 2001: The Social Networks of High and Low
      Self-monitors: Implications for Workplace Performance. Administrative
      Science Quarterly, 46, March, pp. 121-146.
Smith ,E.R. – Mackie, D. M. 2001: Szociálpszichológia. Osiris, Budapest
Smith-Lovin, L. – McPherson, J.M. 1993: You Are Who You Know: A Network
      Approach to Gender. In England, P. (ed.): Theory on Gender/Feminism on
      Theory. Aldine de Gruyter, New York, pp. 223-251.
Stokes, J. – Levin, I. 1986: Gender differences in predicting loneliness from social
      network characteristics. Journal of Personality and Social Psychology, 51, pp.
Utasi, Á. 1991: Az interperszonális kapcsolatok néhány nemzeti sajátossága. In
      Utasi Á, (ed.): Társas kapcsolatok. Gondolat, Budapest
Wagner, W. G. – Pfeffer, J. – O’Reilly, C.A. 1984: Organizational Demography and
      Turnover in Top-Management Groups. Administrative Science Quarterly, 29,
      pp. 74-92.
Wasserman, S. – Faust, K. 1994: Social Network Analysis: Methods and
     Applications. Cambridge University Press, Cambridge
Wellman, B. 1985: Domestic Work, Paid Work, and Net Work. In Duck, S. –
     Perlman, D. (ed..): Understanding Personal Relationships. Sage, London
Wharton, A.S. – Baron, J.N. 1987: So Happy Together? The Impact of Gender
     Segregation on Men at Work. American Sociological Review, 52, October, pp.