208
Lithuania the level of education predicts only two types of co-residence
trajectories, namely Transition to parenthood, which is more likely for
individuals with higher degrees, and Solo type of trajectory, which is less
frequent for those with a medium level of education.
The Influence of Co-residence Trajectories (1990–2010)
on Personal Networks Across the Three Countries
In the previous section, we identified the main types of co-residence tra-
jectories and their sensitivity to birth cohort, gender, and the level of
education of respondents, globally and for each country separately. We
highlighted the importance of birth cohort, as the window of observation
is not the same for both cohorts. We now move to the main aim of this
chapter, which is to conceptually and empirically bring together
trajectories and personal networks as different components of social con-
figurations (Elias 1994 ).
In Chap. 5 , we mapped the diversity of personal configurations and
uncovered nine types of arrangements that were named according to the
main feature(s) of their composition, namely, in order of importance:
Narrow-nuclear (20.2%), Parents (14.3%), Friendship (11.1%), Extended
conjugal (11.0%), Standard-nuclear (10.8%), Sibling-oriented (9.3%),
Work-oriented (8.9%), Beanpole (6.9%), Mixed (5.2%), and Alone (2.2%).
As our goal is to understand the influence that co-residence trajectories
exert on personal networks, we carried out nine regression models in
order to estimate the impact of biographical, structural, and macro-level
factors on the composition of personal networks. As biographical factors,
we retain the typology of co-residence trajectories; as structural factors,
we include birth cohort, gender, and level of education; and as macro-
level factors, we consider the country of residence (cf. Table 7.4). In a
second step, we compute logistic regressions for each country separately.
The corresponding results are described below, but not presented in a
table.
First of all, we can read from Table 7.4 that co-residence factors are the
most impactful ones. Indeed, all nine configuration types are predicted
by at least one type of co-residence trajectories, while it is only six times
J.-A. Gauthier et al.