Participants and non-participants. Problems of representativeness in deliberative democracy experiments

In comparison to former deliberative experiences with self-selection of participants, recent experiments are characterised by a notable effort to increase the representativeness of the participants’ sample. Sampling and recruitment have therefore assumed a significant role for the success of a deliberative event and the external validity of the results of the dialogic-deliberative process. Despite the attention to the selection of participants and the use of control systems based on the main socio-demographic variables, it can be highlighted that people who accept to participate in a deliberative experiment show important differences if compared with individuals who refuse to take part in the same event. These differences can significantly affect the qualitative results of the experiment. The paper examines the participants’ profile of two deliberative polls held in Italy (Citizens’ consultation of Turin) and Europe (Europolis) between 2007 and 2009. After an assessment of participants and non-participants’s features by means of descriptive statistics, the study proposes the application of an experimental model for reproducing a representative sample of the target population and replicating previous results. The objective of the paper is to describe behavioural and attitudinal differences in the real sample, composed of effective participants, and the hypothetic sample, composed of individuals weighted and stratified on the basis of non participants’ socio-demographic and political characteristics. The use of such an experimental strategy is used to distinguish differences both before and after the deliberation experience.