GROUP-SIZE AND THE LOGIC OF COLLECTIVE ACTION - A NETWORK ANALYSIS OFA SWEDISH TEMPERANCE MOVEMENT 1896-1937

In this paper we address the free-rider problem from a network perspec tive. We suggest that individuals' groups of relevant others are consi derably smaller than is usually assumed in the Olsonian tradition. Ins tead of focusing on the interest group as a whole, we argue that a gro up of relevant others consists of those to whom the individual is tied through various social bonds. Since these groups tend to be small, so cial selective incentives are likely to be efficient in inducing indiv idual participation. In testing these ideas empirically, we use microd ata on members of a Swedish temperance movement organization during th e period of 1896-1937. We estimate how individuals' groups of relevant others are composed with respect to membership in the movement organi zation and how the composition of the groups affects individuals' deci sions to join the movement organization. The results of the analysis s upport our thesis that additional movement members in the group of rel evant others increase an individual's propensity to join a social move ment organization. However, the results also lend support to Olson's f ree-rider thesis: When controlling for the composition of the group of relevant others, additional members in the movement as a whole decrea se an individual's propensity to join the movement.