Deliberative Poll

A deliberative poll is a methodology for measuring public preferences that combines small group discussions and traditional scientific polling. It was created by James Fishkin, political science and communications professor, with the goal of improving the quality of public opinion expression and measurement.

Fishkin argues that traditional polls often do not provide good measures of public opinion because members of the public are not knowledgeable enough about the important issues of the day and do not have the motivation or opportunity to engage in deliberation on the issues. He first proposed the idea of deliberative polling in 1988 as a corrective. Fishkin, who has since trademarked the term Deliberative Poll, currently conducts deliberative polls through the Center for Deliberative Democracy at Stanford University.

Typical deliberative polls have three main stages. First, a traditional public opinion poll is conducted of the population of interest, for example, all voting-age adults in the United States. A probability sample of this population is selected and respondents, who agree to participate in all the stages, are asked standard survey questions on selected issues along with some background and demographic questions. Respondents are then sent briefing materials that provide information about these same issues. In the second stage, respondents travel to a given location to deliberate on these issues. The deliberations take the form of small group discussions and can include sessions where participants are able to question experts. Some more recent deliberative polls have used online deliberations. In the third stage, the participants are interviewed again using traditional survey techniques to see whether their views changed as a result of their deliberative participation. Fishkin's view is that this second survey shows what public opinion would look like if the entire population were more informed and able to engage in deliberations on these issues.

The first national deliberative poll in the United States (called the National Issues Convention) was conducted in Austin, Texas, in January 1996, at a cost of about $4 million. A second National Issues Convention was conducted in Philadelphia, Pennsylvania, in January 2003, which was followed by the first online deliberative poll. Some utility companies in the United States have also used deliberative polling at the local level to get public input on energy policies. Deliberative polls have also been conducted internationally in such countries as Australia, Britain, Bulgaria, China, Denmark, Greece, Italy, and Northern Ireland.

Some public opinion researchers have raised scientific concerns about deliberative polling. One challenge is getting a representative sample of survey respondents to participate in the deliberations. In the 1996 National Issues Convention, older respondents, those with less education, and the less politically active were less likely to travel to Austin for the [Page 185]weekend of deliberations. However, selection differences were less prevalent on the issue questions.

Another concern is whether group discussions are the best approach for disseminating information. Deliberative poll participants generally take the group discussion task seriously, but criticisms have been raised about the quality of the discussions and the accuracy of information exchanged in them. A related criticism of the discussions is the potential impact of group dynamics. In group situations, people can be influenced by normative factors unrelated to the strength or merits of the arguments. In addition, differences in discussion participation rates can also have an impact on opinions. Not everyone is equally motivated or has the same ability to participate in group discussions. The more vocal and persuasive members of the group may have a disproportionate influence on the outcome of the deliberative poll.

...