Sex differences in social networks
of overlap, from the perspective of the total number of friends and the total number of group
members, may provide different insights into sex differences in social networks and friendship and
different implications for social development. In this study it was predicted that there would be a
greater overlap between social networks and friendship relations for girls than boys.
The fourth issue addressed in this paper concerns the stability of social networks and
friendships. Because girls’ activities are more centred on intimacy and friendship than those of boys,
we might expect their networks to be more stable over time. Little research has studied network
stability but, where examined, the use of different methodologies and definitions have led to
ambiguous results (see Cairns et al., 1998). Findings suggest high stability over a few weeks and
lower stability over the longer term, especially between years and school transitions. In the current
research we studied the stability of both social networks and friendships in relation to sex and
predicted that girls’ networks and friendship relations would show greater stability over time.
The Structure of Social Networks
A further contribution of this paper relates to the method of identification and analysis of the
structure of networks. Most studies rely on self or peer report involving a top down approach to
social network identification. The peer report approach assumes that children have similar
knowledge, based on informal observations, about social network membership (Kindermann, 1996)
and network identification involves examining the consensus among pupils about which members
co-exist in the social network. This approach is not without its problems (Gest, Farmer, Cairns &
Xie, 2003) and children may not have full knowledge of network membership, particularly in
relation to the opposite sex, and some networks and individuals may be more salient than others. An
alternative method that identifies social networks 'bottom-up' on the basis of repeated and
aggregated systematic observations of children interacting on the playground and conducted by
trained observers, provides a more accurate view of social network membership and structure.