Legislators and policymakers typically refer to public preferences as the basis for setting crime-control and penal policies; however, this assumes that their assessments of informed public opinion are accurate. This paper argues that only by institutionalizing opportunities for reflective and deliberative interchange of ideas between experts and the public can appraisals of the public's will be accurate and useful as input for policymaking. This requires providing a forum for people to express their current views while exposing them to new information relevant to the issues being addressed. The public must then be allowed to analyze the impact of this information on their policy preferences, to reconcile their ambivalent attitudes, and to take responsibility for the consequences of their opinions. The concept of the deliberative poll (DP), which has been proposed by James Fishkin and others, would facilitate and measure such informed public opinion. DP methods are based on open and informed dialog between members of the public and persons who provide accurate and up-to-date research findings on the issues being addressed. Under such a format, respondents are given the opportunity to receive such information, consider its impact on their existing policy opinions, and then decide whether to change or revise their opinion. The author recommends that DPs be conducted regularly, possibly as part of the British Crime Survey; at key stages in the policymaking process, possibly as part of the consultations that inform government White Papers; and on significant occasions, possibly after high-profile crimes that alarm the public. 89 references