119Identifying the Main Shaping Factors of the Types
of As-Family Networks: Regression Analyses
In order to confirm and compare the effects of the main factors which
shape the type of as-family networks, we also ran three logistic regression
analyses (Table 4.9). We considered the three variables: country (Model
a), the birth cohort as a life stage variable (Model b), and gender (Model
c).
The Couple type of as-family network is associated with both country
and gender but it is important to underline that the effect of country is
transversal to all three models. Individuals who were born in Switzerland
are four times more likely to build up this type of network than individu-
als born in Portugal. Individuals born in Lithuania are also nearly twice
as likely to have this type of network when compared to individuals born
in Portugal. Regarding gender, women are less likely than men to present
this type of family network.
The Friendship-origin type of as-family network is predicted by both
country and birth-cohort. Individuals born in Lithuania are less likely
than the Portuguese to have a Friendship-origin type of family network.
However, when the birth cohort enters the model, we can see that the
latter variable is the strongest predictor of this type of as-family network
since the cohort effect (Model b) increases the R^2 from 0.045 to 0.132.
Those who were born in the younger cohort are three times more likely
to build this type of network in comparison with the oldest cohort.
Gender has no statistically significant effect.
The No-family configuration is predicted by the three factors.
Individuals both in Switzerland and in Lithuania have a higher chance of
having a No-family network than individuals in Portugal. Cohort and
gender effects are also significant. Individuals born in the younger cohort
are less likely to not consider any of the members of their personal
configuration as family. In contrast, women are nearly twice as likely as
men to not consider any of their network members as family.
In contrast to the significant effects of country on nearly all the other
types of as-family networks, the Siblings type is only predicted by birth-
cohort and gender. Individuals who were born in the younger cohort are
Changing Meanings of Family in Personal Relationships...