178
the three countries. They also show whether configurations which mostly
include non-kin are characterised by a bridging structure of social capital
within each country.
For density of emotional support, as well as for density of interactions,
the three countries show the same link between family and bonding social
capital, although at different levels (especially for Portugal, which has
higher bonding social capital in almost all configurations than Switzerland
and Lithuania). In all national contexts, configurations where family ties
are under-represented feature a lower level of density, indicative of weak
bonding social capital.
Figure 6.2 shows the centralisation of emotional support across per-
sonal networks in the three countries. Switzerland and Lithuania share a
common pattern: non-family oriented configurations, in their case, pro-
vide significantly more bridging social capital than family-oriented ones,
especially those focused on children.^3 Portugal, by contrast, shows almost
no variation of bridging social capital according to the composition of
personal networks. This lack of relationship between bridging social capi-
tal and the composition of configurations in Portugal is striking, even
when we turn to configurations composed of both kin and non-kin ties.
As shown in Chap. 4 , this is accounted for by the higher integration of
non-kin in the family realm, meaning that there is a higher proportion of
friends and other non-relatives who are considered as family (see Chap.
4 ). In other words, friends in Portugal are very likely to be actively
0%10%20%30%40%50%Portugal Switzerland LithuaniaFig. 6.2 Centralisation of emotional support by configuration, by country (%)
E.D. Widmer et al.