Chapter 13- not spoken to the other for a long period
- shouted at the other
- hit/slapped the other
- thrown something in anger.
To obtain our measure of conflict we add up these replies.
We have measures for ages 2, 3, 6, and 12, and at each age
we measure the average conflict up to that age— or up to
separation if separation has already occurred.
To measure whether the parents are separated, we have
annual reports from the mother on any separation up to age
13, plus the child’s report on separations at ages 13– 16. So at
each age our variable measures whether the original family
is no longer intact.
The Effect of Family Conflict
In analyzing the effect of these variables, it is, as so often, dif-
ficult to know what to hold constant. There are obvious is-
sues on how to handle mental health in particular, since it is
such a powerful determinant of life- satisfaction. Mental ill-
ness may cause family conflict, in which case it should be
included in the equation if we want to find out the spe-
cific effects of conflict without picking up the effect of a
confounder. On the other hand conflict may cause mental
illness as a mediating variable,^3 in which case if we want to
find the effects of conflict we should exclude mental illness
from the equation.
In Table 13.1 we look at outcomes at age 16 following
both approaches. In the top two rows we do not control for
the mental health of the parents. The effects of conflict are