Home |  site-map |  indonesia |  indonesian-single-source-articles |  asteroid-user-registration |

Roy Morgan OZ Panel Login  |  Mail Diary Panel Login |   careers |  contact-us |

It is not generally known that, in the last few years, Gordon Heald of Gallup London and some of our other Gallup International affiliates, have conducted public opinion research in Russia, the Eastern Bloc, and mainland China. (See Appendix A).

These public opinion surveys meant Mikhail Gorbachev and his fellow leaders in the Eastern Bloc were able, for the first time, to really know how their people felt. Awareness of this public opinion knowledge enabled (or forced) them to bring about the changes we are seeing today, which has led them to democracy.

In an interview with Sally McMillan, for The Australian, before the recent Federal election, I said "Now there's democracy in Russia, Australia should be next".

Her article in the February 26 Australian, did not include my quote. However, there is little doubt it still applies except it should read "Australia must be next".

Opinion polls and freedom of information are as essential a part of a democratic system as the freedom of speech.

Of course as with any freedom, freedom of information carries with it responsibilities:

responsibility on the part of the pollster to conduct and report the research findings honestly and accurately;

responsibility on the part of the journalists to report the research correctly, in context, and even-handedly; and

responsibility of political parties and the Government to use this information properly, (not for agenda setting by alluding to private poll results which are never authenticated, and not misusing the information in polls).

Today I have several critical points to make about the misuse and abuse of public opinion polling in Australia and about its role within the democratic process.

However, before making those points let me say we, The Roy Morgan Research Centre take our responsibility very seriously. As well as the moral and ethical issues involved, our reputation depends on it.

Using public opinion polls to predict election results is the main way the accuracy of market research companies can be assessed. In this regard, Roy Morgan Research's Morgan Gallup Poll has an unrivalled record of accuracy.

The documentation attached (See Appendix B) shows clearly that the Morgan Gallup Poll was the most accurate in predicting the 1987 and 1990 Federal elections, and the last NSW, Victorian, South Australian and Queensland elections.

Of course Roy Morgan Research does not only conduct the Morgan Gallup Poll, we are the largest Australian market research company.

We have, since 1972, conducted the industry standard readership survey. With our joint venture partner Dun and Bradstreet, A.C. Nielsen (Australia) Pty. Limited is now establishing in every mainland capital city daily TV ratings based on TV meters.

In the last 12 months we have sold survey data to the four major political parties and during the recent Federal election the Liberal Party and the Australian Democrats commissioned political surveys from us.

It is a concern that intelligent people such as Alan Ramsey (SMH-November 25, 1989), Jennifer Beacham (Press release-March 18, 1990), Bob Hogg (AM-March 22, 1990 and National Press Club-April 11, 1990) and Leonie Kramer (The Australian, April 30, 1990) would question the accuracy of the Morgan Gallup Poll. While I'd suggest that Leonie Kramer's comments are more along the line of devil's advocate, Bob Hogg's comments, when considered in the context of his behaviour over many years as Secretary to Victorian ALP and Federal ALP, clearly had political overtones, as did Alan Ramsey's and Jennifer Beacham's.

It would not be unreasonable for the average voter to ask why Bob Hogg spends so much of his time criticising Gary Morgan and the Morgan Gallup Poll. (Jennifer Beacham and Alan Ramsey should be put in the same category as Bob Hogg).

If we look at the issue in perspective, in its true context, the "politicking" is obvious. The attacks (and they were coherent, well documented, thoroughly researched although wrong!) on the accuracy of the Morgan Gallup Poll came in the last few days in the run-up to the Federal election - at the time when the nightly Morgan Gallup Polls provided strong evidence that the Multi-Function Polis issue (kept alive by the ALP no doubt on the advice of Bob Hogg and Rod Cameron) had backfired and was costing the ALP significant support.

Because of ALP propaganda, people are led to believe it was the Victorian State ALP that nearly cost Mr Hawke his job.

"Of the two sentiments expressed, both of which I heartily applaud, one was reflected in the votes in Queensland and northern New South Wales, by the reaction to corruption in public life when exposed. The other, expressed most strongly to our cost in Victoria, was that at some point Government, in this case the State Government, must accept the political responsibility for their perceived or real failings. To duck that responsibility guarantees a rebound on the party." (Bob Hogg's address to the National Press Club - April 11, 1990)

However it was, in fact, the Federal ALP's handling of the MFP issue which almost cost the ALP Government.

Clearly there is no doubt Andrew Peacock raised the MFP issue. It began as an attack on the ALP. Most business leaders, academics and politicians would be in favor of greater international exchange of technical knowhow from overseas, although not if such an exchange involved a Japanese "enclave".

The issue was kept alive by Laurie Oakes when he spent the first 5 minutes of his March 18 "Sunday" interview with Andrew Peacock on the MFP. The attack on the ALP's support for the MFP would have finished then if either Senator Button or Prime Minister Hawke had referred to the Federal Government's specific rejection in 1987 of any prospect of a Japanese "enclave".

It seemed strange at the time that Senator Button didn't make this point and clear the air. He had in fact said there would be no Japanese enclave on February 14, 1990 at a Committee for Melbourne meeting which included members of the Victorian State Cabinet and Victorian business leaders.

From Andrew Peacock and the Liberal Party's point of view, Peacock's initial attack on the MFP did not help politically - criticism was levelled at them for being racist. The Liberal Party would no doubt have preferred to let the issue die a natural death.

However, Rod Cameron (who has recently resigned yet again as ALP's pollster) and Bob Hogg had no intention of letting the MFP issue go away. They made the strategic political decision to use the MFP issue to show division within the ranks of the Liberal Party, ie. that John Elliott and Andrew Peacock were divided. (It is ironic that John Cain made available to the press the John Elliott to him letter dated August 29, 1989 approving of the MFP.)

The MFP issue continued in Monday's Australian with Paul Kelly's article "Peacock a 'danger in the Lodge' " and Monday night's ABC TV news lead with John Elliott, Hugh Morgan, Nobby Clark and Brian Loton (all members of the Committee for Melbourne) being shown to be against Andrew Peacock's stand on the MFP. Rod Cameron was well aware of the ABC TV news story before it went to air. (Is it democracy at work when the chief advisor to the ALP election campaign is made aware of details of the ABC TV lead news story hours before it goes to air?). In fact, Rod Cameron had been busy on the telephone on the Monday afternoon (before the ABC TV news story) preparing the way for the "ALP launch of the MFP issue".

Unfortunately for the ALP by Tuesday the issue wasn't the MFP, it was immigration* and the ALP was losing support. The Morgan Gallup Poll conducted on Tuesday night had the Liberals in front for the first time. It was obvious to us that the immigration issue was the issue which put the L-NP in front in the last week.

*Note: During 1989 both the ALP and Liberal Party purchased the results of the continuous Roy Morgan Political Issues Surveys. Both parties were well aware of the September 1989 result that 18% of electors believed reducing the number of migrants to be one of the three most important Federal political issues. On February 18, a month before the election, the Melbourne Sunday Herald released the February Roy Morgan Political Issues results showing the immigrant issue rated equal third with reducing personal income tax, and after reducing interest rates, and protecting the environment. This Political Issues poll was taken before there was any mention in the media by either political party that immigration was an issue.

In the May 13, 1990 issue of the Melbourne Sunday Sun we released the results of a poll conducted on May 10 which clearly showed Australians against the MFP (52% said it was a bad idea, only 27% a good idea, with 21% undecided). The poll also showed Australians were strongly in favor of reducing or stopping the number of migrants from anywhere (69% favored reducing or stopping immigration, 27% opposed reductions, with 4% undecided). Greatest opposition to the MFP was from Victorians where 59% said the MFP was a bad idea (in the Federal election the ALP suffered most losses in Victoria).

For Bob Hogg and his colleagues to defend themselves by arguing that the late swing against the ALP (particularly in Victoria) was not due to their own tactical error in playing up the MFP issue, but rather a spurious poll result, ("the Morgan swing") is understandable - but ridiculous.

To suggest that the Cain State Government was the cause of the swing against the ALP in Victoria is equally ridiculous.

Before Bob Hogg's onslaught on the Cain Government, the Morgan Gallup Poll found the Victorian ALP had recovered support. No doubt this was due to the Cain Government's handling of the transport dispute, which achieved a major change in work practices. (John Halfpenny believes poor management of public transport is a key reason for the Cain Government's decline in popularity - See The Australian, June 20). The Morgan Gallup Poll showing an improved vote for the Victorian State ALP (and therefore proving Cain and the State ALP were not responsible for ALP Federal loss of support) was vigorously attacked by Bob Hogg as being inaccurate.

There is little doubt that the Cain Government's later fall in popularity, and Mr Cain's fall in personal approval, can be attributed to the internal bickering started by Bob Hogg and his colleagues after the Federal election.

It's ironical that John Cain has, in the past, fought so hard to protect details of nearly $1 million worth of polling data supplied to his Government (see page 10) by Rod Cameron's ANOP when Cameron and Hogg were involved in the handling of the MFP issue which nearly cost the ALP the Federal election.

There is no doubt that there are very real and potential dangers in the misuse and abuse of political polls. However, before getting involved in that area it's worth making clear exactly why public opinion polls are an essential part of a democracy.

I'd like to quote what my father said just before he died in 1985, when asked about founding public opinion polls in Australia in 1941*.

"Why was measuring public opinion important?"

"Firstly, to stop journalists from claiming to be able to write 'The public demands, etc.' when they don't know what public opinion is on that particular subject.

* Interview David Jones had with Roy Morgan just before he died. From manuscript: "Roy Morgan, The Gallup Poll Man".

Secondly, Socrates stated the central 'core' of a problem as a question. This forces people to face the problem. Public opinion polls not only do this, but also make sure that (by publication) the public and journalists know the problem and the public's attitudes towards the issue (as percentages). This forces our leaders and the public to recognise the real problem and hopefully consider it.

Thirdly, publishing several times a year reasonably accurate percentages for voting intentions creates a continuous 'election atmosphere' rather than, a phoney atmosphere created by so called 'leaders and journalists'".

Although my father's references were to political polling in the 1940's, his thoughts are just as relevant today when measuring public opinions. He, of course, had not then experienced political parties playing the same games as journalists and claiming knowledge of the public demands, etc., and alluding to polls which no-one could check.

Just before my father died he said he was concerned that political parties were beginning to learn how to use results from public opinion polls to help them quickly change the political agenda during election campaigns.

There is little doubt that the leaders of the Queensland National Party believed the Morgan Gallup Poll, published 11 days before the November 1986 election, which showed a large decline in their support. This information was the catalyst for the significant changes in the National Party's campaign strategy. This was not the first time poll results were the catalyst for such changes.

In 1975 Don Dunstan (then the Premier of South Australia) used polls, unfavourable to the ALP, as the excuse for attacking the then unpopular ALP Whitlam Government only days before the South Australian election. The ALP Dunstan Government was returned.

During the 1980 Federal election the Liberal and National Parties completely changed their campaign when all polls showed the ALP well in front. A telephone poll conducted by the Morgan Gallup Poll on the Wednesday before the election was the only poll to predict accurately the election result of a Liberal-National Party win.

Until the November 1986 Queensland election, the Morgan Gallup Poll had, since 1946, correctly predicted the winning party for each election surveyed. The Queensland election remains probably the best example of how a political party can, after seeing results of a political poll, change their complete strategy and successfully turn around what seemed to everyone to be inevitable failure. (In 1986 the Morgan Gallup Poll did not survey in Queensland during the week before the election.)

When a political party believes that the unfavourable results of a public opinion poll are accurate, then not changing their campaign strategy would be committing political suicide.

Of course it is not acceptable if polls are the only input into policy and/or strategy. There are situations where public opinion must be ignored. For instance, if a political party has a strong commitment to a publicly unpopular policy or strategy (eg. fixing the economy by improving productivity through changes in work practices such as penalty rates, working on Sunday, working on rostered days off, etc.) it is essential the party maintains its stance (true to its belief).

But what about the real dangers of polls, and the suggestion that they should be banned, (In the March 25 Sunday Age it was reported that Rod Cameron "believes opinion polls are a disgrace and serve only to increase the cynicism of the electorate". He was quoted as saying "I don't like banning things, but it's sorely tempting to act as dictator and ban opinion polls") or the suggestion from a question at Bob Hogg's Press Club address on April 11 that opinion polls be subject to monitoring or auditing by the Australian Electoral Commission.

There are any number of real life examples of abuses of public opinion polls.

1) In December 1982, the leaking of Cameron's ANOP poll results on Hawke versus Hayden was undoubtedly the catalyst which instigated the move to have Hayden stand down as the Opposition leader. No actual evidence (eg. computer sheets) of this ANOP poll result has ever been made available. In hindsight, an approval rating of 38% (as found by the Morgan Gallup Poll for Hayden in December 1982 and January 1983) is not especially low for an Opposition leader.

2) In the run-up to the 1988 NSW State election Cameron's ANOP research was reported in the Sydney Morning Herald as showing that Mrs Greiner's drink driving charge was a major reason why people would vote against the Liberals. This was clearly a blatant manipulation of the truth, and an abuse of Cameron's position of authority as an unbiased pollster. Understandably, Barry Unsworth and Steven Loosley were outraged when they found ANOP's misinformation obtained more publicity than the Premier's policy launch!

There is little doubt that Steven Loosley is well aware of Cameron's history of providing misleading information based on fictitious poll data.

3) In 1988 in Victoria it is obvious that, contrary to what Rod Cameron claimed, the ALP called the Victorian State election when they believed they had a large lead in the polls.

When the election was announced the Morgan Gallup Poll showed ALP support at 51%, Liberal Party at 32%, National Party at 8% and 9% for others. (See Appendix C)

4) Also in the 1988 Victorian State election, given the ALP's carefully targeted campaigning in marginal seats, the swing against the ALP was remarkably consistent across Victoria (with a few exceptions such as the most marginal seat of Bentleigh). ALP propaganda released by Rod Cameron and reported widely in the press on the Monday immediately after the Victorian election (eg. Age-October 3, Herald-October 3), claimed the ALP gained support in the marginal seats. An assessment of the electoral office's polling data shows the ALP actually lost support in the marginal seats (except Bentleigh) as they did everywhere else. (See Appendix D)

5) Cameron and the ALP are, of course, not the only offenders. In the recent Federal election campaign Andrew Peacock and Tony Eggleton continually claimed their private polls showed the L-NP ahead in the key marginal seats, and that the L-NP would win by 16 seats. It is interesting that this Liberal claim was not criticised by Cameron or the ALP. Strategically, it seems, such propaganda suited the ALP. (The ALP preferred public image coming up to an election seems to be "slightly behind the Opposition").

Because most electors believed the ALP would win the Federal election, Peacock's widely publicised claims of a 16 seat lead in private Liberal Party polling only seemed to further diminish his already poor credibility.

6) In the May 5 Age Norman Abjorensen in his front page story on John Elliott made the following reference to Liberal Party research:

"Early in 1989 private Liberal Party research began to identify Mr Elliott as a liability. It showed repeatedly that he was regarded as being the strident voice of big business, of representing, at a time of declining living standards, interests already more than well-heeled."

Norman Abjorensen gave no source to verify his statement. Indeed in early 1989 John Elliott's standing as the Liberal Party President was near its highest! (See Morgan Gallup Poll Finding 1836)

7) And it's not only politicians and party strategists who abuse public opinion poll data. Westpac and the Institute of Applied Economic and Social Research at the University of Melbourne have since the beginning of this year made spurious comparisons between their new telephone survey which purports to measure Consumer Confidence and the long running series of data collected by Roy Morgan Research Centre by door-to-door interviewing - resulting in misleading claims that Australians' Consumer Confidence is at an all time low. (See Appendix E)

Roy Morgan Research's results, which are strictly comparable with the historical series of 17 years, has for the last three months shown Consumer Confidence higher than June to July last year. As released in the June 15 BRW the June Roy Morgan Consumer Confidence Index is 83.1. Consumer Confidence will drop if people like Westpac/Melbourne Institute continue to release wrong information. (See Appendix F)

This wrong information was given credibility by being quoted on June 4 in the London Financial Times as "a record low in the respected Westpac-Melbourne Institute index of consumer sentiment, which fell 2.7 points in May to 70.6".

"Lies, damn lies, and statistics" has been a catch cry for many years now. But it hasn't caused us to stop people counting, or to outlaw numbers, or to take any other such draconian, anti-democratic action.

Control of abuses of public opinion polls needs to follow a scientific model - with guidelines for quality, and openness to inspection of results and methods (with of course provision for the maintenance of commercial and intellectual confidentiality rights), rather than a model of State control - with an official poll, whether it be monitored or regulated by the Electoral Commission, the Australian Bureau of Statistics or the Government of the day.

Two separate letters written by me and published in the March 1988 Times on Sunday and November 1989 Bulletin set out very clearly the issues of responsibility which need to be considered in relation to the use of polls for setting the political agenda. (See Appendix G)

No value can be placed on public opinion poll results unless the reliability of the data is beyond dispute. No reference should be made to such public opinion poll results unless backed up with specific details of:

• When the survey was conducted,
• Where the survey was conducted,
• How the survey was conducted,
• Actual questions asked, and answers given,
• Sample size, and
• Who paid for the survey (it seems that in some instances political surveys have been paid for by Governments).

It's understandable that political parties try to present poll results in the best possible light.

It is up to the journalists not to simply publish "propaganda", but to check and publish the important details, and the actual facts.

A self-regulatory system whereby pollsters and journalists (and perhaps even politicians) behave responsibly in relation to this important issue of freedom of information may be hard to manage, but is infinitely preferable to State control.

We already have enough control - Australia is too much the country of big business, big Government, and big unions. We don't need a big official, bureaucratic polling regulatory body, which authoritatively reports the public's opinion (with the "public" having little or no knowledge of the results nor recourse to contradict, question or check that "reported opinion").

Nor do we need Government paid, regulated and controlled political advertising.

However, I want to be clear, I'm not saying smaller Government is all we need. Competition in all areas of business (Government, companies/businesses, and workers) is the single most important ingredient. Unfortunately there is no political party in Australia today who will face up to this fact.

In a recent survey of 1,000 British academics, business and Government economists, few believed in the efficiency of privatisation as such: 80% believed privatisation would not raise cost efficiency unless coupled with measures to increase competition.

Bob Hogg's recent suggestion of the Government paying for political advertising is of course another attempt to curtail the future of our democratic system - a system which requires freedom of information, communication and therefore advertising. While Bob Hogg is flying this agenda it should be remembered that a High Court challenge was needed to stop the Hawke Government increasing Government's funding on politicians' postage. The postage allowance is used for political promotion paid for by the Government. However there are no controls on telephones being used by Members of Parliament (both State and Federal) for party purposes. It is well-known that this practice occurs. State and Federal Governments should release details on the amount each Member of Parliament spends on the telephone. Electors would be astonished if the political parties released figures on how much they spend using direct mail to communicate with electors in the marginal seats.

So much for the moral and ethical issues - the suggestion of Government paid political advertising raises another very real question. Who pays for all this? Of course it's us - the taxpayer.

When the market forces are working correctly - public opinion research is undertaken when it is believed that the benefit of knowing the results of the research outweighs the cost, and the benefits of a particular methodology, considered in conjunction with the cost, represents value for money.

The market forces don't work naturally in Government - and if decisions about public opinion research (or any Government expenditure) are not based on a correct assessment of the real benefit-cost ratio the public pays too much.

It has, until recent years, been very difficult for research companies, other than Cameron's ANOP, to win contracts from many Government Departments (many research companies were not even invited to tender).

Canberra today seems to be becoming more professional with the principle of lowest tender price being accepted. However, there are still many Statutory Authorities and Corporations, heavy spenders of taxpayers' money, who don't have to adhere to this principle.

In a recent Telecom contract for $11,700,000 The Roy Morgan Research Centre was advised the following by Telecom on the suggestion that the lowest tender hadn't been accepted:

"It is not Telecom policy to disclose accepted prices or the difference in price between that accepted and that of the unsuccessful tenderer. I am therefore unable to provide any additional information in this respect."

In a democratic society it is essential that such information be available - and open to public scrutiny. Apart from the fundamental issue of freedom of information - not making such information available creates suspicion that, in this case, either Telecom didn't accept the lowest tender or Telecom changed the tender specifications and didn't in fact call tenders on the new specifications. In either case, the existing contract must be cancelled.

I won't go into this in more detail except to say a considerable amount of evidence on this topic was given to the Joint Parliamentary Committee of Public Accounts - Engagement of External Consultants by Commonwealth Departments. The minutes of evidence are available.

However, as pointed out earlier during 1984 to 1987 Cameron's ANOP was the recipient of nearly a million dollars worth of surveys from the Victorian ALP Cain Government. Not only was this expenditure of taxpayer's money not made with "value for money" as a priority, but as of today, no computer printouts or the specific questionnaires used by interviewers have been made available to prove that ANOP actually conducted those surveys or to establish the reliability of their data.

It would be interesting to know whether Bob Hogg and/or John Cain have seen the questionnaires and computer printouts of the following surveys:

Some of the Surveys conducted for the Victorian Government

by Cameron's ANOP - 1984 to 1987

In Canberra during 1984 to 1989 Cameron's ANOP was also the recipient of over a million dollars worth of survey contracts, many given by Mick Young's Department. Some of those surveys were:

© 2012 Roy Morgan Research. All Rights Reserved
privacy-statement