This article analyses the impact of decision-making methods on democratic deliberation. An experiment was designed to study the effects of two distinct decision-making methods. This citizen deliberation experiment took place in November 2006. The topic was: ‘Should a sixth nuclear power plant be built in Finland?’ A random sample of 2,500 people was first polled and invited to take part in the deliberation experiment. Eventually, 135 people took part in the event where they read written information, heard an expert panel and deliberated in small groups. The participants were randomly divided into two treatments. In the first treatment, the small group decisions were made through a secret ballot. In the second treatment, each group was asked to formulate a common statement reflecting the group's opinion. The present article analyses changes in the participants' opinions and knowledge, as well as indicators of social pressures. There were no systematic differences between the treatments in terms of opinion changes but the participants' knowledge increased more in the common statement groups. When it comes to social pressures, the results show no systematic differences between the treatments.