110
of respondents citing female friends excluded them from the family net-
work. In contrast, in Portugal and Lithuania, the drop is far from this
percentage, with a difference of only 3% (male friends) and 8% (female
friends) between the friends cited as important and those who are consid-
ered as family. This points to a clear-cut difference between friends and
family in Switzerland, in contrast to an assimilation of friends as family
in Portugal. In Lithuania, the drop is not as striking either, but it is
important to recall that in this country the percentage of friends identi-
fied as family is close to zero.
Thirdly, although step-family members are not highly represented in per-
sonal networks, they are systematically considered as family. This is transver-
sal to the three countries as we see a full convergence between the percentage
of respondents who cited step-family members as important close persons
and the percentage of respondents who considered step- family ties as family.
In other words, when post-divorce family relationships are recognized as
important, the process of suffusion between family and step-family tends to
be strong. As all three countries have high levels of divorce, these results also
confirm the increased acceptance of post- divorce family relationships in
these countries and in European societies in general.
In summary, this first look at those persons considered as family within
personal networks shows that the mechanisms underlying family mean-
ing continue to be strongly influenced by blood and alliance principles,
in particular those related to the nuclear family, to co-residence over the
life course and to emotional support (overall in this order of importance).
However, in Portugal and Lithuania, blood and alliance principles seem
to go more easily beyond the central family dyads, including more ties in
descending and ascending line, while in Switzerland there is a stronger
focus on the traditional components of the nuclear family of procreation
(partners and children). Another interesting finding is that emotional
support, although important, does not emerge as the overriding mecha-
nism of family meaning within personal networks. It is more important
in Portugal (where nine in ten persons considered as family provide emo-
tional support), but even there it increases only slightly when compared
to the number of alters providing support in personal networks. In con-
trast, the proportion of as-family ties who lived in the same household as
ego (co-residence) increases more substantially in the three countries.
K. Wall et al.