Sex differences in social networks
17
and number, reported earlier, yet highlight the important social-relational implications of socialising
in social networks of different sizes.
Social Network Stability
Table 3 shows data for the stability of cores and groups over the year according to strict
(100% stability) and more liberal criteria (51% or more stability). None of the male cores or groups
were 100% stable. However, nearly two fifths of girls' cores and a quarter of girls' groups were
completely unchanged over the year. Using the more liberal definition, approximately equivalent
proportions of male and female cores were stable. All of the boys’ groups were moderately stable
while a third of girls’ groups were unstable. These stability patterns, however, may be influenced by
the size of the network and thus favour boys. Estimates of stability not affected by network size are
based on the proportion of an individual’s relational connections within groups at T1 that were still
present at T2 (see Table 4). These show that on average boys’ network relations were more stable
than those of girls (F(1,105)=8.17; p< .01, d = .57).
In contrast to the data for social networks, best friend and friendship relations were unstable
(see Table 4). Girls’ friendship, but not best friend, relations were more stable over time than those
of boys (F(1,123) = 4.03; p< .05, d = .36).
Discussion
This study was concerned with sex differences in the size and structural composition of the
social networks of 7-8 year old children arising from observations of interactions during breaktimes.