When making decisions, MPs may seek to discern public opinion on the issue.

Opinion polls and surveys

Governments frequently seek out the public’s opinion on a range of issues.  Often what they seek is ‘public opinion’ in its most basic form – the wide range of views that randomly selected individuals have on any given issue.  This is the most common use of public opinion and it is captured in a couple of ways:

• An opinion poll is a series of questions asked in a standard way to a ‘representative’ sample of the population.  These polls are normally conducted by telephone or in person.
• A sample survey is similar to an opinion poll but involves a more detailed questionnaire.  These surveys are most often used by academic researchers and the government and provide more detail than opinion polls.

Polls and surveys are useful means of bringing to the public realm the voices of ‘ordinary’ people between elections.  They also allow the governed to tell the government their thoughts on particular issues.  This is not easily inferred from a person’s vote on election day where we grant a mandate to govern.  Polls also allow politicians to test the ‘public mood’, thus keeping them in touch with the public between elections.  In this sense opinion polls and surveys ‘oil the wheels’ of democracy.

But polling has its limitations.  Polls are blunt instruments best suited to simple questions and ill equipped to gain insights on more complex issues.  They often do a disservice to the public’s opinion by oversimplifying it.  In addition, the political elite decides what questions will be asked in polling and therefore essentially frames the debate.  One of the strengths of representative democracy is that it uses deliberation and debate to enrich the elected representatives’ knowledge on issues.  Opinion polls, on the other hand, do not allow the respondent to discuss the issue before expressing an opinion.  Nor do they ensure the respondent is educated on the issue before commenting.  For these reasons Sir Robert Peel, twice elected British Prime Minister, described public opinion collected in this way as ‘the great compound of folly, weakness, prejudice, wrong feeling, right feeling, obstinacy and newspaper paragraphs’.  The complaint is that public opinion polls merely tally up many individual, often ignorant, viewpoints.

The more ambitious definition of public opinion is the informed judgement of a community on an issue of common concern, where that judgement is formed based on shared political goals.  By this definition, public opinion is more measured, considered, informed and communal.  It is not enough to gather together individual opinions; public opinion must be shaped in the cauldron of collective debate and discussion.  Opinion polls and surveys are, by this broader definition, unable to effectively define public opinion.

Deliberative polling

If this richer vision of public opinion is accepted then more sophisticated techniques of ‘polling’ are needed.  There is growing support for the use of deliberative polling.  This technique involves picking a small sample of voters and exposing them to a variety of positions on an issue, including presentations from experts and politicians, to ensure the sample group of voters is knowledgeable on the issue before them.  Only then does the group debate the issue and reach a decision.  Opinion is only measured after the issue has had a thorough airing, thus making the deliberative opinion poll a means of polling informed opinion.  According to James Fishkin, ‘an ordinary opinion poll models what the public thinks, given how little it knows.  A deliberative opinion poll models what the public would think, if it had a more adequate chance to think about the questions at issue.’  

Citizens' juries and citizens' assemblies

Citizens’ juries and citizens’ assemblies are examples of deliberative opinion polling because the aim is to take a microcosm of the public and allow them to deliberate and then make a fully informed decision.  The citizens’ assembly is a larger citizens’ jury.  The key differences are the number of people who are involved and the time taken to hear evidence and deliberate.

The idea borrows many basic components from the jury in a court: people are selected at random, presented with evidence and asked to deliberate in order to make a decision on a specific and pre-specified question.  Random selection is important to ensure that the jury is a microcosm of the wider society, and so its decisions should be accepted as representing the views of all.  Provision of evidence is a crucial component of the process and usually includes written reports and access to experts with a range of specialisms and perspectives.  The importance of deliberation is emphasised by the use of professional moderators.  The decision is communicated in a written report and will contain a recommendation on the question asked and some explanation.  Again, professional assistance is provided to assist the process but one of the aims is to have the report written in language that is accessible to the general public.

When the jury or assembly is created the legislative body will specify what it intends to do with the report.  It may treat the report in the same way it would treat a report from a Royal Commission or Inquiry: as a set of ideas to be considered and debated with a written response and explanation for actions taken, or not taken.  It may state that the recommendation will automatically be put to a vote, either of the legislative body or by the population in a referendum.  The legislative body may state that it will follow the recommendation.

A jury has between 12 and 20 members, whereas an assembly has over 100.  A jury usually meets daily for no more than two weeks, while assemblies meet less often but take up to a year to complete the process.  Citizens’ assemblies have been used for constitutional issues, in particular electoral systems, while juries are often used to settle questions in relation to policy implementation, such as where to place a halfway house for people who have left prison, or how to ration river usage.

In British Columbia, a citizens’ assembly was made up of 160 randomly selected British Columbians – 80 men and 80 women – representing the province’s 79 electoral districts.  There were also two First Nations members.  Their mandate was to review different electoral systems, to determine whether the province’s existing FPP system should be retained, to recommend an alternative electoral system if not, and to examine the likely impact of this recommended system on the province’s system of government. Participation in the assembly was voluntary; members received reimbursement for expenses, including child care, and a $150 honorarium for each day the assembly was sitting.

The assembly first met in January 2004. Its work was divided into three phases:

• Learning about electoral systems: Members met over a series of six alternate weekends in Vancouver when they learned about and studied electoral systems in use around the world.
• Public hearings: The assembly held 50 public hearings all over British Columbia and sifted through the 1,603 submissions made by the public.
• The deliberation phase: From September through November members deliberated and worked towards their final report. They recommended a form of Single Transferable Vote (STV). [link to electoral systems]

The recommendation was put to British Columbian voters in a referendum which coincided with the May 2005 provincial election. Legislation required a double majority for the referendum to pass and for electoral reform to be implemented: 60% of voters in 60% of the province’s electoral district. This was a high hurdle; the referendum failed to pass.

The province of Ontario followed a similar process. Its citizens’ assembly has recommended MMP.

Referenda

A referendum allows voters to have a direct say on an important policy or constitutional reform. It is an instrument of direct democracy often used to supplement the legislature in representative democracies. Referenda can be proposed by the government or the people and the results can be consultative or binding.

Most referenda are proposed by the government but a few countries allow citizens’ initiated referenda (CIRs). New Zealand is one of these countries, alongside Switzerland, Italy and most states within the USA. In each case there are rules about which issues can be proposed, how many signatures need to be gathered in a petition and the time allowed for the gathering of signatures. In New Zealand a petition needs signatures from 10% of eligible electors and organisers have a year in which to collect them. The result of the vote is not binding on the government. There have been three CIR votes.

Consultative referenda are a means for the government to test the mood of the public, because the government makes no commitment to follow the majority view. Binding referenda are used as instruments of ratification: a government will run a referendum to gain final public approval for a decision taken concerning a very important issue such as constitutional reform. Most countries have had at least one referendum but few use them a lot. In Australia, the Republic of Ireland and Switzerland any change to the constitution has to be ratified in a referendum.

In New Zealand the issue of the electoral system was voted on in a consultative and then a binding referendum. In 1992 voters were asked two questions: if they thought the electoral system should be changed; and which of four alternative systems they thought should be considered. This was a consultative referendum but the government chose to follow the results and so held a binding referendum in 1993 which gave a choice between the existing FPP system and MMP.

Those who argue for the use of referenda see them as a way for the people to have their say rather than relying upon the elected representatives. Those who are wary of the use of referenda prefer to have elected representatives make the decisions and in particular are worried that voters will vote emotively rather than having considered the evidence, will be swayed by well-financed campaigns and may be overtly conservative or discriminatory in their decisions.