In this paper we distinguish between two measures of contact between c
hildren and their non-residential fathers: whether or not the children
have seen their father in the previous year, and how many weeks the c
hildren stayed with their father. The distinction between the two is i
mportant because there are many aspects of a parent-child relationship
, some of which may be met during shorter contacts, and others which m
ay be nurtured during an extended visit at the father's house. Using d
ata from the 1987-88 National Survey of Families and Households we fin
d that having a legal agreement regarding joint custody is positively
related to the number of weeks that a child stays with his/her father.
Children are more likely to see their fathers if there is a legal chi
ld support agreement and the father pays child support, which indicate
s that paying child support and seeing the child are complementary act
ivities. This pattern is reversed for the number of weeks the child st
ays with the father, which gives support to the hypothesis that non-re
sidential fathers substitute one activity for another. This research h
ighlights the need for careful consideration of different measurements
of contact between non-residential parents and children in order to u
nderstand more fully the dynamics of families following divorce.