Network Density and Virtual Social Capital of Israeli Adolescents: Network Density and Virtual Social Capital of Israeli Adolescents Ilan Talmud and Gustavo S. Mesch Department of Sociology University of Haifa


Two Theories of Social Capital: Two Theories of Social Capital Bridging Social Capital Ego enjoys social disconnectedness Non-redundant ties Sparse networks Low Density Binding Social Capital Ego enjoys trust, emotional support Strong ties Transitive relations Cohesive networks


Slide3: Cohesive Network Structural Holes


Binding Social Capital: Binding Social Capital Putnam’s network connection Nahapiet and Ghoshal’s corporate social capital Wellman’s network social capital


Bridging Social Capital: Bridging Social Capital Granovetter’s Weak Ties Burt’s Structural Holes Integration: Linking Social Capital The Emergence of Network Society (Castells, 1998; Wellman et all, 2000) Multi-cultural and multi-ethnic societies (Uslaner, 1999; Putnam, 2000)


Two Approaches to Social Capital: Two Approaches to Social Capital Normative: civic virtue, trust, identity, Structural: relations, cohesion, homophily, incentives forced by structural arrangements Not mutually exclusive, but rather complementarities


Off-Line and Computer-Mediated Social Networks among Israeli Youth : Off-Line and Computer-Mediated Social Networks among Israeli Youth Users and Non-Users Attributes and Relations Binding Social Capital and Network Density


Recent Studies: Recent Studies Negative relationship between age and Internet use, as the frequency of use is higher among adolescents (Kraut et al., 1998; Mesch, 2001). Addiction (Brenner, 1997; Griffith, 1999) Social isolation (Kraut et al., 1998), Lack of involvement in pro-social behavior (Funk and Buchman,1996), Poor school performance (Van Schie and Wiegman, 1997).


Sample: Sample A nationally representative, stratified random sample of the adolescent population of Israel. 1000 respondents, February - April 2001.


Does Internet Connection Reduce or Increase the size of social ties? Average Number of Friends: Does Internet Connection Reduce or Increase the size of social ties?   Average Number of Friends


Descriptive Findings: Descriptive Findings The number of friends that were named as close friends does not differ between adolescents who are connected and adolescents who are not connected to the Internet Adolescents who are connected to the Internet, report slightly more friends than adolescents who are not connected. Yet, this difference was not found to be statistically significant. The sample of adolescents who are connected is more heterogeneous than adolescents who are not connected


What Proportion of Adolescents’ Ties are Internet Friends?Proportion of Friends that Were Met on the Net (For Internet Users Only): What Proportion of Adolescents’ Ties are Internet Friends? Proportion of Friends that Were Met on the Net (For Internet Users Only) Between 7.1 and 13.2 percent of all the friends named were met through the Internet. For the first friend, 13.1 percent of the sample reported that this particular friend was met through the net, for the second friend 9.5 percent reported that the friend was met for the first time through the net. There are no noticable differences by gender


Length of Acquaintance for Internet Users and Non Internet Users: Length of Acquaintance for Internet Users and Non Internet Users On average, the duration of knowing a face-to-face friend is longer than knowing duration of an Internet friend For non-Internet users, the closest the friend, the longer the duration of knowing him/her. For Internet users, the more distant the friend, the longer the duration of knowing him/her.


Where do the friends of Internet users and non-users live?Distribution of Place of Residence of Each Friend Named by Internet Use : Where do the friends of Internet users and non-users live? Distribution of Place of Residence of Each Friend Named by Internet Use   In general, a high percentage of the friends live in the city and a small proportion in the neighborhood. the real difference appears when it comes to friends overseas. Internet Users report more friends overseas, whether they meet them in the net or just keep their relations with them online.


Distribution of Place of Residence of Internet and Face- to- Face Friends (Percentage) : Distribution of Place of Residence of Internet and Face- to- Face Friends (Percentage) The majority of both Internet and face-to-face friends of Israeli adolescents are locally based. When he/she is a face-to-face friend 51.6 percent live in the respondent neighborhood and 41.6 percent live in the same town. Internet friends have a higher tendency to live outside the respondent's town and county.


Resources Exchange What resources are being exchanged by adolescents and their friends through the Internet?: Resources Exchange What resources are being exchanged by adolescents and their friends through the Internet? The closer the friend, the more likely are information and software to be exchanged. As the friend was named latter, the nature of exchange tends to be more social and the less instrumental. Software and computer resources are exchanged by about a fourth of the sample.


Trust and Closeness to Internet Friends: Trust and Closeness to Internet Friends For Internet users, there is a difference if the friend is one that was known by internet or face to face. This difference is only noticeable for the first friend, in which the adolescents report more closeness, importance and trust, comparing to a friend they met via the Internet. For the closest friend, trust and closeness are a result of meeting face to face.


Network Social Density: Network Social Density Ratio: Number of contacts / Possible contacts % of Alters who know one another Transitivity Measured for Users and non-Users


Slide19: Pairwise Correlations among Network Properties


Network Density by Gender : Network Density by Gender Girls’ network density is slightly higher than boys’, though statistically not significant. ♀♂


Slide21: Network Density by Nationality There is an average difference in network density by nationality. Adolescent Israeli Jews have lower network density than Arabs (α < .02), indicating that friends of Israeli Jews tend to know each other less (=less Transitivity).


Slide22: A. Non Internet users B. Internet users Density By Nationality, Controlling for Internet Use Among non- Internet users, there is a significant difference according to nationality. Israeli Jews have significantly less dense networks than their Arab counterparts. Yet, this deference is less pronounced among those adolescents who are connected to the Web.


Network Density by Religion: Network Density by Religion Network Density significantly varies by religious denomination This pattern holds even after controlling Intenet Linkage


Zero-Order Correlations between Density and Tendency to Ask for Help from Various Sources in Stressful Situation: Zero-Order Correlations between Density and Tendency to Ask for Help from Various Sources in Stressful Situation The more adolescents belong to a dense network of friends, the more likely they are to rely on friends or on a Rabbi. The findings may indicate that individuals that belong to a dense networks will rely on them under most circumstances .


Slide25: Tendency to Seek Help Varies Across Religion


Slide26: Approaching a Source of Potential Help is Affected by Social Attributes * p<.05 ** p<.01


Importance of Friend and Density(correlations): Importance of Friend and Density (correlations) The findings show that, overall, there is a greater and significant tendency to assign importance to social contacts as network density increases


Average Strength of Relations* by Internet Connection (*Importance of a contact, Duration, Closeness, Possible commitment of a contact, and Trust in a contact ; Cronbach's α = .95).: Average Strength of Relations* by Internet Connection (*Importance of a contact, Duration, Closeness, Possible commitment of a contact, and Trust in a contact ; Cronbach's α = .95). Strength of relations was lower among those who are linked to the Web (α <.05).


Strength of Relations by Gender : Strength of Relations by Gender Strength of relations is significantly higher among girls (24.3), in comparison to boys (22.98) (α < .05).


Strength of Relations by Nationality : Strength of Relations by Nationality While total relational strength is not significantly different, the average strength of relations is significantly higher among Jews (24.1), in comparison to Arabs (21.5) (α < .05).


Predictors of Relational Strength : Predictors of Relational Strength As network density increases, strength of relations decreases. Internet connection is negatively associated with relational strength.


Predictors of Internet Connection (logistic) : Predictors of Internet Connection (logistic) As network density increases, strength of relations decreases. Among the attribute variables, nationality has the most dramatic effect and gender has an additional significant net effect.


Conclusions: Conclusions Ego’s Binding Social Capital is negatively associated with ego’s network density and Internet linkage Network social density is associated with a person’s attributes (demographics) Internet Connection is associated with network structure and ties properties Problem of aggregation between ego’s social capital and network (group’s) social capital


More Research is Needed: More Research is Needed Further research using longitudinal or panel designs is needed to examine how social networks are changing via virtual and face-to-face relations. Examining more specifically rival hypotheses regarding the sources of network choice (i.e. structural constraints vs. culture). Studies of adults virtual and off-line social capital.