The importance of participation in voluntary formal associations for e
nhancing health is supported by four kinds of evidence. (1) Seven pros
pective studies of social relations and mortality show the independent
effect of formal social participation, net of informal ties. (2) Conc
eptual analysis demonstrates that voluntary formal associations consti
tute a separate class of social causation. (3) Our factor analysis of
data from a sample of 629 nonmetropolitan elderly identified two types
of formal social participation: ''instrumental,'' as in associations
that are community oriented, and ''expressive,'' as in those that exis
t for the benefit of the members. (4) Regression analysis showed that
the instrumental participation factor is linked, net of controls, to t
he perceived health of both men and women, whereas the expressive fact
or predicts for women only.