Too Many Friends: Social Integration, Network Cohesion and Adolescent Depressive Symptoms

Abstract

Using a nationally representative sample of adolescents, we examine associations among social integration (network size), network cohesion (alter-density), perceptions of social relationships (e.g., social support) and adolescent depressive symptoms. We find that adolescents with either too large or too small a network have higher levels of depressive symptoms. Among girls, however, the ill effects of over-integration only occur at low levels of network cohesion. For boys, in contrast, the ill effects of over-integration only occur at high levels of network cohesion. Large social networks tend not to compromise positive perceptions of friend support or belonging; whereas, small networks are associated with low perceptions of friend support and belonging. Hence, perceptions of social relationships mediate the ill effects of under-integration, but not over-integration, on depressive symptoms. Roughly 30 percent of adolescents report moderate to severe depressive symptoms (Rushton, Forcier and Schectman 2002). The early occurrence of depression in adolescence sets a foundation for recurrent and severe depressive episodes later in life (Belsher and Costello 1988; Kovacs et al. 1984). Depression in adolescence is also an urgent health concern. Depressive symptoms are the strongest predictor of suicidal ideation which, in turn, predicts suicide attempts (Kandel, Raveis and Davies 1991). Suicide is the fourth leading cause of death among 10-14 year olds in the United States and the third leading cause of death among 15-24 year olds (Anderson 2001). This research explores how the network structure and perception of adolescent friendships influence depressive symptoms in adolescence. Several decades of research make a clear link between social relationships and depressive symptoms in adolescence. This is not surprising given that cultivating and managing peer relationships is a central developmental task of adolescence, requiring much time and energy. The vast majority of research on peer relationships focuses on perceptions The authors gratefully acknowledge funding from the William T. Grant Foundation. We would also like to express thanks to Jim Moody for graciously sharing his SAS Programs for Analyzing Networks users manual.

Psychological Variables in Researches between Social Network and Social Capital

“…Then, Zhou, Shin, Brass and Choi [10] found a curvilinear relationship between the creativity of employees and their weak ties. The individual network analyzed by Falci and McNeely [11] showed a curved linear relation between the size of friend-network and depressive symptoms of teenagers.…”

Adolescent Social Networks: General and Smoking-Specific Characteristics Associated With Smoking

“…The IP does not assess the extent to which participants are socially integrated (central) versus isolated (peripheral) in their social networks' structure. Likewise, the IP limits participants to nominating a maximum of 10 network members, and adolescent networks are undoubtedly larger than that (Falci & McNeely, 2009). The measure was also limited to assessing cigarette use and did not assess use of other tobacco products, such as e-cigarettes.…”

Adolescent Time Use, Companionship, and the Relationship with Development

“…Researchers also have examined the structural properties of social networks (Falci & McNeely, 2009 ;Haynie, 2001Haynie, , 2002Krohn, Massey, & Zielinski, 1988 ;Schreck, Fisher, & Miller, 2004 ) . These network properties are derived from respondent friendship nominations and describe the connections between an adolescent and the other respondents' nominations, such as how many nominations a respondent receives (popularity), the density of the network, the degree to which members interact with each other in different contexts (multiplexity), or the degree to which a respondent is embedded in the network (centrality).…”

“…This indicated that whether an adolescent's friends were delinquent mattered more for popular than less popular youth when predicting whether an adolescent was likely to engage in delinquency. Other network research examples include support for the relationship between increased multiplexity and decreased adolescent smoking (Krohn et al, 1988 ) , a decreased risk of violent victimization for youth in dense, conventional networks (Schreck et al, 2004 ) , and higher levels of depressive symptoms in very large or very small networks (Falci & McNeely, 2009 ) . Kreager et al ( 2011 ) recently expanded previous research by examining how group-level characteristics affect network properties.…”

“…While most school-based surveys include a majority of nominations that are to other youth attending the same school, there can still be a sizeable number of out-of-school nominations. For example, an examination of the Add Health data found that 68% of friendship nominations were made to same-school adolescents (Falci & McNeely, 2009 ) . Given the problematic outcomes associated with out-of-school friends and the demonstrated importance of peers, an obvious bene fi t to research would to be able to study these nonschool youth who may be instrumental in negatively affecting developmental trajectories of in-school adolescents, such as in fl uencing a decision to drop out of school or join a gang.…”