Sex differences in social networks
18
The study examined whether social network size was related to the type of games played, the nature
of the emergent internal structure of male and female networks, the overlap between friendship, best
friendship and social networks, and the extent to which networks and friendships were stable over
time.
Social Network Size
Consistent with previous research was a sex cleavage in the composition of social networks
and evidence that female networks were greater in number, while male networks were on average
larger (Belle, 1989; Benenson et al., 1998). To evaluate competing hypotheses regarding the
connection between team game play and social network size, we examined the size of male and
female play networks observed on the playground when playing all games, non-team games and
team games. Consistent with the ‘two worlds’ model, male larger play networks were due to the
playing of team games because when the team game data were removed from the analysis there was
no difference in play networks size. However, examination of the social networks from the
aggregation of the play network data for non-team games only, showed that these were of equivalent
size to those including the team game data. This favours the notion that there is a ‘centripetal force’
that draws boys together into large networks (Benenson et al., 1998). It therefore appears that large
networks are sustained whether children engage in team games or not. The possibility remains,
however, that an interest in team games establishes the boundaries of the social network, which is
then maintained in other types of games. A number of other factors also point toward a ‘team games
determine network size’ interpretation. Existence of a second male network in every class that was
comparable in size to those of girls suggests that the ‘centripetal force’ model does not relate to all
boys. Researchers’ notes indicated that networks had relatively distinct game profiles with some
networks engaging in primarily fantasy play while others involved more competitive/athletic, social
and cooperative activities. Finally, other research suggests that networks increase in size from the
early years and then decrease in size during adolescence (see Cairns et al., 1998; Pitcher & Shulz,