The United Kingdom’s first Deliberative polling was implemented in 1994 in order to better gauge and then impact public opinion on crime reduction. The results indicated that policy stances did change as a result, particularly among highly-educated participants.
Problems and Purpose
Since 1981, when The Crime Survey for England and Wales (CSEW) began their first survey, crime in England and Wales has been rising consistently. Crime reached its peak in 1995. [1] The first Deliberative Polling in England and Wales looked to gauge public opinion on the rising level of crime during and leading up to 1994.
In an effort to seek a solution to rising crime levels, the first Deliberative Polling in England and Wales looked to see the effect of deliberation and the deliverance of information regarding crime to see if it had an impact on the opinion on participants from the United Kingdom. The techniques employed by the Deliberative Polling method and the results obtained, look to help policy makers in the United Kingdom create policy to tackle rising crime levels.
Background History and Context
The method of the Deliberative Poll was the creation of Professor James Fishkin of Stanford University. First revealed in 1998, the method was contained within the article; ‘The Case for a National Caucus: Taking Democracy Seriously’. [2] The Deliberative polling method was defined and constructed in greater detail a year later within ‘Democracy and Deliberation: New Directions for Democratic Reform’. [3]
Organizing, Supporting, and Funding Entities
The deliberative polling method was a registered trademark of the Center for Deliberative Polling at the University of Texas at Austin. James Fishkin is the director of the organisation (At the time of the creation of this entry, the Austin Centre has moved to Stanford as of September the 1st, 2003 and operates as the new centre for Deliberative Polling). The organisation receives funding from the use of the trademarked method. The funds raised support the research conducted by the Center. [4]
The Deliberative poll was co-sponsored by both channel 4 and the Independent newspaper. The Independent newspaper had an extended roll of involvement which included providing the moderators used for the general running of the poll. [5] The Kettering foundation provided the training to the moderators in line with the standards expected by the Centre for Deliberative Polling. In addition to this, the Kettering foundation guided the creation of the briefing materials that were given to participants. [5] The Kettering foundation is a non-profit organisation which has a focus of co-operative research on democracy. Founded in 1927, the organisation looks to answer the question, ‘what does it take to make democracy work as it should?’. [6] The Kettering foundation had already had experience with organising citizen discussion of public policy via its own global network of group-based discussions known as National Issues Forums. [6] The sample of participants was selected by the Social and Community Planning Research Institute (SCPR), an independent research institution based in London. The SCPR also played a large roll in recruitment and the collection and preparation of data. [7]
A large proportion of analysis was done by Robert C. Luskin while being funded by the University Research Institute of the University of Texas at Austin; Luskin is the Research Director of the institute. Luskin was given access to the facilities by the Centre d’Etude de la Vie Politique Franc¸aise of the Fondation Nationale des Sciences Politiques, Paris. [5] A proportion of the procedure was video taped and produced by Granada Television and shown on British television by the broadcaster, Channel 4 television. [5]
Participant Recruitment and Selection
Participant selection consisted of a random sample of the electoral roll, therefore meaning that the sample was taking from all registered UK voters. [7] This random sample was the product of a three-stage probability sampling design from all registered UK voters from south of the Caledonian Canal. Stage one consisted of a selection of forty parliamentary constituencies; the probability of a constituency being selected was proportionate to the number of registered voters within that constituency. [5] The selection was stratified by region, percentage of owner-occupiers, and population density. Stage two followed, which was to randomly select a polling district within each consistency (probability was also parallel to the number registered voters). Finally, stage three had a simpler process of a random selection of voters from each polling district, with equal probability. [5]
In addition to the selection of participants, members of UK parliament were selected to sit a plenary question-and-answer session. The three candidates selected represented the 3 biggest political parties in the UK at the time of the study. The candidates consisted of the Chancellor the Exchequer: Kenneth Clarke (Conservative), Gordon Brown (Labour) and Malcom Bruce (Liberal Democrat). [8] These additional participants took questions towards the end of the weekend thus meaning that participants were able to ask questions that had been formulated from deliberation from across the weekend. The enticements offered to participants consisted of an all-inclusive trip to Manchester, with food, travel and a ‘good’ hotel being offered in exchange for participation. A monetary offer of £50 was also given for any who decided to participate. [5]
Methods and Tools Used
The method used for this particular case study is the Deliberative Polling method. This method encompasses a collection of tools and techniques used to produce the results in this poll. Due to the Deliberative Polling method being trademarked, the tools and techniques that are implemented are monitored by Center for Deliberative Poling at the University of Texas at Austin, in order to ensure consistency among the cases. [5]
Tools and Techniques
A national random sample was used in this poll. Described in detail in the previous section, the random sample intends to draw a group of participants that are representative of the wider electorate. After the sample was selected, a pre-deliberation survey occurred consisting of face-to-face interviews coupled with self-administered questionnaires. The various array of tools and techniques that follow are implemented after the first initial questionnaire. Therefore, the impact of these tools and techniques cannot be distinguished from one another. Any possible effects caused by these tools are techniques are recorded with the final questionnaire at the end of the study.
Small group discussions consisting of randomly assigned groups were a key component along with a question and answer sessions where the political elite and policy experts took questions by participants.
What Went On: Process, Interaction, and Participation
The ‘pre-deliberation survey’ was the first instance of the tools and techniques being used in this method. This initial stage occurred with 896 respondents; a response rate recorded at 74%. [5] This component consisted of a 15-minute interview segment followed by a self-administered questionnaire that culminated in a final interview section. Finally, the respondent was asked to participate in the Deliberative poll. To help retain interest in the poll among the selected participants, regular phone calls occurred to encourage and retain commitment among all participants. During the review of the study, Fishkin admits that 301 participants that agreed to participate in the poll were, ‘self-selected’. [8]
Influence, Outcomes, and Effects
An analysis of the results conclude that there was change in policy stance in all groups. The highest average change in policy attitudes were exhibited by groups who had the greatest levels of education, at a value of 16%. Participants who were less educated exhibited an overall change of 5%. The Deliberate poll aimed to combine the deliverance of information with deliberation among participants to influence a change in the overall stance of a particular policy. Fishkin argues that these two tools and techniques were the catalyst for policy attitude change. [7] The fact that the more educated succumbed more to policy change suggests that participants were not being influenced solely by new information; there was a process of deliberation that causes the change, otherwise the less educated would have been more adept to policy attitude change as well. Fishkin states that the more educated were more willing to weigh up competing arguments during deliberation, thus claiming that the deliberation was a source of policy attitude change. [7] With the poll being televised, the process and results were able to distributed across the UK to provide greater awareness of the Deliberative Polling Method. The UK’s first deliberative poll was also used during network coverage of the 1997 election as decided by the British government. [7]
Analysis and Lessons Learned
The Deliberative Polling method was designed to provide a possible solution to the common flaws found in modern public opinion polling methods. Fishkin claims that the primary hindrance to the polling method is high levels of ‘citizen-ignorance’ relating to a lack of credible information when asked an individual is asked to take stance on a particular issue. [3]
Referred to by Robert Dahl as minipublics (minipopulus), the Deliberative poll utilizes a minipublic to create a random sample of the electorate. Dahl states that the ‘demo should include all adults’ who are ‘subject to the binding collective decisions’. [9] A common issue faced by the creation of minipublics is who out of the population is recognised as being bound by the decisions of this minipublic. [9] The electorate in this Deliberative Poll is the chosen population sample. Issues with solely using participants from the electoral roll is highlighted by Ian O’Flynn and Gaurav Sood. [10] The homeless, prisoners, hospitalised and active-duty soldiers are subject to a binding collective decision by a deliberative poll but are noted as have a near 0% chance of inclusion. However, Flynn and Sood note that the random sample does ‘stand of firmer ground’ than other methods of sampling such as quota sampling. The stratification of the sample also mitigates the possibility of the loss of segments of the population. [10] Issues concerned with the attrition of the sample are highlighted with specifics to Britain’s First Deliberative poll by Luskin and Fishkin.
The effect of inaccessibility and self-selection is an issue that remains passive during the recruitment of participants. [11] However, the recruitment was noted by Luskin and Fishkin as being ‘perfectly representative’, despite the issues of attrition. Out of 102 items of sociodemographic groups, and policy attitude clusters, only 14 were found to be statistically significant differences. [11] With Fishkin arguing that the cause of possible attribute changes in relation to policy are down to deliberation (referred to in the previous section), Jon Elster states the possible issues of the deliberation among small groups; possibilities for ‘skilled and charismatic speakers’ to dominate discussions who ‘count on rhetoric rather than argument’. [12] This potential aspect of deliberation within small groups is pivotal to the validity of the study, as Elster would suggest. This is particularly relevant to the UK’s first deliberative poll as there was a high number of small group discussions across the weekend. The use of moderators used during the discussions who were trained by the Kettering Foundation, still do not remedy this effect according to Everett Carrl Ladd, who claims that this particular poll ‘neglects lessons from decades of research on group dynamics’. [13] Ladd states that the group discussions are unrepresentative of actual political discussions because of a certain set of social interactions and information flows. [13]
See Also
References
[1] Office for National Statistics (2018) Crime in England and Wales. June 2018 Edition. London: Office for National Statistics. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/bulletins/crimeinenglandandwales/yearendingjune2018 / [Accessed: 17/11/2018]
[2] Fishkin, J., S. (1988) The Case for a National Caucus: Taking Democracy Seriously. Atlantic Monthly, August, 16-18.
[3] Fiskkin, J., S. (1991) Democracy and Deliberation: New Directions for Democratic Reform. United States of America: Yale. Yale University Press
[4] Center for Deliberative Democracy (2018) What is Deliberative Polling. Avalibile from: https://cdd.stanford.edu/what-is-deliberative-polling/ [Accessed 04/12/2018]
[5] Luskin, R., C., Fiskkin, J., S. and Jowell, R. (2002) Considered Opinions: Deliberative Polling in Britain. British Journal of Political Science, 32, 455-487
[6] Kettering Foundation (2018) About. Available from: https://www.kettering.org/about [Accessed 04/12/2018]
[7] Fishkin, S., J. (1994) Britain Experiments with the Deliberative Poll. The Public Perspective. August, 27-29
[8] Luskin, R., C., Fishkin, J., S., Jowell, R. and Park, A. (2000) Learning and Voting in Britain: Insights from the Deliberative Poll. Stanford: Center for Deliberative Democracy. Available from: https://cdd.stanford.edu/mm/2000/general_election_paper.pdf [Accessed: 01/12/2018]
[9] Dahl, R., A. (1970) After the Revolution: Authority in a Good Society. United States of America: Yale. Yale University Press.
[10] L., O., Flynn and Sood, G. (2014) What Would Dahl Say?: An Appraisal of the Democratic Credentials of Deliberative Polls and Other Mini-Publics IN: Grönlund K., Bächtiger, and Setälä, M. (eds). Deliberative Mini-publics. Untied Kingdom: London. ECPR Press. 41-58
[11] Luskin, R., C. and Fishkin, J., S. (2005) Deliberative Polling, Public Opinion, and Democracy: The Case of the National Issues Convention. United States of America: Texas. University of Texas at Austin.
[12] Elster, J. (2012) Deliberative Democracy. United States of America, New York. Cambridge University Press.
[13] Ladd, E., C. (1996) Fishkin’s ‘Deliberative Poll’ is Flawed Science and Dubious Democracy. The Public Perspective. January, 41-44
External Links
https://cdd.stanford.edu/mm/2000/general_election_paper.pdf