Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
The political significance of opinion polls

Download PDFDownload PDF

Department of the Parliamentary Library

RESEARCH NOTE Parliamentary ResearchService Number 43, 7 May 1996 ISSN 1323-5664 The Political Significance of Opinion Polls Introduction Opinion polls play a significant role in Australian politics. The precise character this takes is a contested issue. Arguments focus on who drives opinion polls: the public, political actors (politicians, political parties and interest groups) or the pollsters. This research note assesses the political significance of opinion polls by reviewing the role of each of these groups.

The Public The claim that popular opinion drives opinion polls appears compelling. The popularity of the Prime Minister or Leader of the Opposition, gauged by opinion polls, remains crucial in debates over party leadership (see chart). Policy is also seen to be shaped by attitudes expressed in opinion polls. For example, political interest in environmental issues has been linked to popular

concern expressed through polls.1

To interpret politics as a response to opinion polls, however,

provides only a superficial

understanding of political events. Opinion polls may play a part in determining a party's leadership, but this is only one concern

among many including the

candidates capacity to lead the party or articulate a cohesive political platform. Similarly, opinion polls do place particular issues on the political agenda, but they rarely shape specific

policies, determine how

effectively those policies are implemented or provide insight into policy areas that remain outside of public debate because they are not of interest to the general public, highly technical or a security risk.

Political Actors The view that political actors drive opinion polls is widely held. From this perspective

political actors are interested in the polls because they provide the information required to sell a political image or particular policies. This has resulted in the common criticism that opinion polls tend to represent elections as a horse race, with candidates being primarily concerned with how they react to data published in the latest polls.2

It is certainly the case that

political actors use opinion polls to assess their public image and identify ways it might be

improved. Moreover, polls that indicate popular reactions to specific issues clearly help

political actors to tailor their political campaigns. For instance, the ALP's emphasis on its support for Australia becoming a

Republic may be interpreted in this way because opinion polls had suggested that this would strengthen the Party's electoral appeal. The view that political actors drive opinion polls is

premised on the notion that the political impact of polls is limited

to how political actors present themselves and the way they publicise or avoid particular issues. Polls are not seen as a way for the public to influence the policy position of the political actors in question.

It is only logical that political actors prefer to use polls for their own purposes. Yet the majority of opinion polls that are the

subject of media analysis (and therefore public interest and debate) are undertaken by

independent polling organisations like Morgan, AGB: McNair and Newspoll, all of which avidly defend their impartiality. This suggests that the majority of opinion polls that enter the public domain are far from being the tools of political actors.

The pollsters The particular role played by pollsters is determined by

whether they are independent or privately commissioned. The activities of independent polling organisations are closely tied to the media whose principal concerns involve a mixture of 'selling news' and a commentary on contemporary political

The relationship between opinion polls and surviving as party leader: Downer's Downfall


1 0

2 0

3 0

4 0

5 0

6 0

M a y J u n e J u l y A u g u s t S e p t O c t N o v D e c J a n


P o pularit

y in

opinion polls (per cent)

Downer's leadership unchallenged

Downer's leadership challenged from within the Liberal Party

Downer stands down after internal party pressure

1 9 9 4 1 9 9 5

debates. These polls tend to be on political questions of public interest, such as party popularity, party leadership, voting

intentions or popular reactions to issues of notoriety. Because the reputation of independent polling organisations is based on their perceived impartiality, the

influence of pollsters is typically restricted to their role in selecting issues and designing the

questions asked in polls, both of which may be seen as introducing bias into independent polls. It is widely recognised that issues can be brought into political debate by forms of media attention such as polls. Thus, polls and the

prejudice of pollsters, can play a role in defining political debate. Bias may also be found in the sorts of questions that pollsters ask. For example, polls

commissioned to investigate public reaction to the ALP

government's Native Title

legislation in 1993-4 varied

considerably depending on

whether or not respondents were informed that Native Title

required Aboriginal people to demonstrate an ongoing

association with the land. Those

polls that failed to provide this information had a clear bias built into the questions.3 Whether bias is intended or not, it clearly has political ramifications, effecting popular perceptions of the issue, media attitudes and political calculations.

When political actors commission firms to undertake polling on their behalf they are adopting a political strategy, one aimed to achieve particular political

outcomes. There is no assumption by either the client or the polling organisation that these polls will be used in a fair manner. For instance, details of poll outcomes are usually only leaked into the public domain when it suits the client. The political significance of this sort of activity varies

depending on how astutely the information that has been

gathered is used. The crucial point is that this sort of polling is designed to gain political

advantage. Therefore, when these polls are released, they need to be treated with a high degree of caution, recognising that their aim is to further the specific

interests of a political actor.


• Some polls do influence

political decision making. This tends to be rare,

however, and most polls do not influence the

implementation of policy.

• Political actors would like to be able to manipulate opinion polls. Their capacity to do so is restricted by the fact that the most widely publicised opinion polls are undertaken by independent polling

organisations whose

reputations are based on their impartiality.

• The role of opinion polls

varies depending on their type. It is therefore important to identify whether an opinion poll has been undertaken by an independent organisation or has been privately


• Independent opinion polls are far less likely to be biased, but this is not always the case and all polls should be analysed with a degree of caution.

Privately commissioned polls are highly susceptible to

manipulation by political actors. These polls should be recognised as a part of particular political strategies and scrutinised


Dr Liz Young Statistics Group Parliamentary Research Service

Phone: 06 2772485 Fax: 06 2772454

Views expressed in this Research Note are those of the author and do not necessarily reflect those of the Parliamentary Research Service and are not to be attributed to the

Department of the Parliamentary Library. Research Notes provide concise analytical briefings on issues of interest to Senators and Members. As such they may not canvass all of the key issues.

© Commonwealth of Australia




a g e o


p e op

le p

o lle


Attitudes to Mabo: Impact of question wording

AGB: McNair Poll

Morgan Poll





Source: M. Goot 1993 "Polls as Science, Polls as Spin" Australian Quarterly , v. 65 (4) p. 137

1 C. Bean, I. McAllister and J. Warhurst 1990 The Greening of Australian Politics: The 1990 Federal Election Melbourne: Longman Cheshire.

2 M. Goot and T. Beed 1979 "The Polls, the Public, and the Reelection of the Fraser Government" In H. Penniman (ed) The Australian National Elections of 1977 Washington D.C, American Enterprise Institute.

3 M. Goot 1993 "Polls as Science, Polls as Spin: Mabo and the Miners" Australian Quarterly v. 65 (4) Summer 1993: p.137.