© The Author(s) 2019. Deliberative democracy is for many the most significant development in democratic theory in the last 50 years and it has been used in some places to solve real-world policy problems. However, measuring the impact of deliberative methods is not clearly achievable because several independent variables are manipulated simultaneously. One of the main goals of deliberation is opinion change and thus it is important to understand why opinions change. This paper utilises comparison groups in order to isolate the impact of deliberation from information in the deliberative process. We outline the results of a quasi-experiment in which deliberation took place in a citizens’ assembly (CA) in Ireland. As part of this we measured the impact using pre- and post-test controls, including a group given the information the CA participants received, but without the deliberation. The results of the analysis provide evidence of a separate deliberation effect.