Sex differences in social networks
Aggregation of observation data across time points in close proximity has been employed in the past
and found to be robust (Martin & Fabes, 2001). Few studies make use of direct observation to
identify social networks, yet there is a need for research to examine how social networks arise from
day-to-day behavioural interactions and processes (Rubin, et al., 1998). Direct observation also
enables analysis of the emergent internal network structure in terms of the differential time members
spend together and the way members cohere into full networks. Peer report approaches cannot
examine emergent network structure except in terms of limited indicators of the salience of
individuals within the network. Systematic observation of pupil social interactions on the
playground was the approach used in this study.
This study therefore seeks to examine sex differences in the size and internal structuring of
social networks, the overlap between social network, best friendship and friendship relations, and
stability in social network membership and friendship. Social networks are identified from the
aggregation of repeated observations of children interacting together during breaktime at the start
and end of a year in school.
Method
Design
Children entering Year 3 of primary school in England were studied longitudinally for one
academic year. Data were collected over 8 weeks at Time 1 (T1), the start of the year (September -
October), and then a further 8 weeks at Time 2 (T2), the end of the year (June-July). The
independent variables were sex and time point and the dependent variables were: size of play
networks during team, non-team and all games and reciprocated peer nominations of best friends
and friends.
Participants
Four classes of Year 3 pupils (7-8 year olds), from four separate schools in London,
participated in this study. Permission to conduct the research was sought from and granted by