125family relationships in the traditional components of kinship in Western
societies, may sometimes be excluded from family meaning. Family net-
works in Switzerland also reveal two other important trends: the Couple
type of network based on partnership is largely overrepresented (one in
eight as-family networks in this country) and respondents are more
restrictive with regard to in-laws, and also to intergenerational relation-
ships in descending and ascending line. In contrast, in Portugal and
Lithuania, there is a tendency for individuals to perceive important per-
sonal ties in general as family and to branch out more frequently towards
intergenerational ties and in-laws; this would point, in these countries, to
a process of overlap whereby close personal relationships are imbued with
family meaning and easily absorbed into the realm of the family. However,
in Portugal, even those that are non-kin tend to be imbued with family
meaning, driven by a strong process of suffusion between kin and non-
kin and between personal and family ties, while in Lithuania the overlap
is mainly driven by the overall salience of kin in personal relationships: in
this country, personal networks are almost fully composed of kin, with
the issue of co-residence (having lived in the same household at some
point of ego’s life) also emerging as crucial.
Secondly, testing for the shaping factors of the main types of networks
largely confirmed the influence of contextual factors and the importance of
considering such social conditions in future theoretical models of personal
and family networks. National context and birth-cohort are of major sig-
nificance while the effects of gender are weaker but also crucial with regard
to some types of family networks. Findings show that national context is a
strong predictor of as-family networks for two main reasons: first, with the
exception of the Sibling and Restricted-nuclear types of network, which
both represent low proportions of the total number of networks, national
context is a significant predictor of all the main types of as-family networks;
secondly, country effects remain significant when we add the other factors.
On the other hand, birth-cohort is also a very relevant shaping factor, in
particular as a significant predictor of the Friendship-origin (younger cohort)
and the Beanpole (older cohort) types of network. Gender is undoubtedly
the less important predictor, although it also has a significant role in the
prediction of the Couple and Restricted nuclear (men rather than women)
networks, as well as the No-family and Siblings (women rather than men)
Changing Meanings of Family in Personal Relationships...