Using the Social Ties. an Ethnographic Analysis of a Snowball Sampling

Modelized by James Coleman and others researchers of the Columbia University, the snowball sampling method is now neglected in the social networks analysis. In particular, this method is very difficult and hard to apply and the non-representativeness of the samples that it produces dissuades from using it. However, this sampling method presents many over kind of resources for ones who want to study social networks. First, when the limits of the groups are not know, it permits to identify chains of relationships in a more precise way than “name generator” devices. Moreover, it gives us the opportunity to understand the logics of exchange and structuration which characterized the individuals and groups studied. A reflexive appraisal of the construction of the snowball sample gives us an access to this kind of information.

The aim of this communication is to present the conclusions of a critical feedback from a field experiment during which three snowball sampling waves were conducted between 2009 and 2012. The experience consisted in a transfer of questionnaires person-to-person within subjects’ social networks. The members of the first sample (N=10) have distributed some questionnaires to some of their relatives, friends, neighbours or co-workers who have done the same, and so on. The reflexive analysis of this sampling procedure shows us: 1) The motivations of the subjects to be involve in an active way in a sociological survey 2) The social logics which have oriented the transfer of the questionnaires.

This ethnographic feedback suggests that snowball sampling, beyond the question of statistical representativeness of the samples, maybe inadequate with the concept of social network, allows us to access to realistic networks data.