The Politics of Collating Public Opinion:
Two Cases of Misrepresentation of Public Opinion in Hong Kong


Robert Ting-Yiu Chung
(Director of Public Opinion Programme, the University of Hong Kong)

Note: This article represents the view of the author and not the University of Hong Kong.




Democracy can be defined in a very simple way as "the rule of the people by the people". In modern societies, democracy invariably translates to mean the choice of leaders by universal suffrage through open and fair elections. According to Schumpeter, the democratic method in its modern form refers to the "institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people's vote."(1942: 269) In other words, democracy in its modern form is achieved when people elect their own leaders through competitive elections.


With the advancement of scientific social surveys, one important feature of competitive elections is the occurrence of pre-election opinion polling before general elections, beginning from the 1940s (Crewe, 1992:478). To put it in a nutshell, pre-election opinion polls emerged when both the people and the leaders-to-be wished to gauge the chance of certain individual or party winning an election, or re-election. They are thus only meaningful in societies where popular elections exist, hence their close connection with democracy. Very often, they shape election campaigns.


Other than predicting election results, opinion polls also serve as a barometer measuring a leader's success in retaining public support between elections, hence his/her chance of winning the next election. In some societies, they are even used by government or its opposition to decide when to call for an election. As history develops, the scope of opinion polls was also expanded to cover the general performance of political parties, government departments, local officials, and political figures who affect the life of the people. Their major function in democratic societies, however, is still to facilitate the change in leadership if and when the people asks for it.


Because sample survey rests on the premises that every citizen has an equal chance to be selected, it matches very well with the egalitarian principle of democracy - that each citizen has an equal chance to participate and an equal voice when participating, the existence of opinion polls based on sample survey is also considered as an indicator of true democracy (Verba, 1995). No wonder why in undemocratic societies, opinion polls are often feared and banned by the government, because their results could be used to challenge the legitimacy of the government. In a recent global survey, 30 of the 66 countries covered "have embargos on the publication of poll results on or prior to election day" (Spangenberg, 2003:1) - and this might just be the tip of the iceberg as many countries did not participate in the survey "and these blind spots are precisely where the political situation makes such restrictions even more likely." Likewise, in societies where people's mandate are not important, there may not be any need for opinion polls even if they could be freely conducted. Such may be the case for Hong Kong.


Before 1980s, under the British colonial rule, Hong Kong's political leaders have never been subject to any popular elections. Then, in the early 1980s, the issue of what would happen to Hong Kong after 1997 became an important issue for Britain and China, followed by rounds of negotiation. It was against this background that, in 1982, the British Hong Kong Government set up some small local councils with very limited jurisdictions, called District Boards, and introduced major direct elections for the first time.


In 1984, after a series of negotiations, Britain and China finally signed the "Joint Declaration of the British and Chinese Governments on the question of Hong Kong" which confirmed the handover of sovereignty in July 1997, and the creation of the "Hong Kong Special Administrative Region" (HKSAR) within the People's Republic of China thereafter. Hong Kong people is guaranteed 50 years of full autonomy under the "one country, two systems" formulation. There is, however, no provision that the top leader of Hong Kong, called the "Governor" before 1997, and the "Chief Executive" thereafter would be popularly elected before 2007. Thus, although electoral politics gradually evolved in Hong Kong, it is still far from becoming a democratic society. In fact, it was as late as 1991 that Hong Kong people became able to vote for a small portion of the Legislative Council, which is the supreme legislative body in Hong Kong.


However, as expected, important policy issues emerge from time to time, which requires people's mandate. In the early 1980s, the sovereignty of Hong Kong beyond 1997 was an important issue for the entire society, while in the late 1980s, whether or when to introduce direct elections into the Legislative Council was the subject for public debate. A few years before the handover, in the mid-1990s, constitutional reform again became controversial, as the last Governor of Hong Kong, Chris Patten, decided to speed up the pace of democracy in spite of China's strong opposition. A few years after the handover, in 2002, the HKSAR Government's proposal to enact anti-subversive legislation to protect national interest sparked massive discontent. In each of these important events, public opinion was at stake.


Under colonial rule, the British Hong Kong Government practiced "absorption politics" whereby public opinion was channeled to the administration mainly through the local elites. Such a system was gradually replaced, or enhanced, by gradual democratic reform since the 1980s, whereby public opinion could, to some extent, also be reflected by popular votes. However, because Hong Kong's top leaders are not returned by popular votes, it remains an important task for the government to demonstrate that it is "responsible" and "responsive" to people's demands.


This paper examines two major incidents in the history of Hong Kong, one under the British rule, and the other under "one country, two systems", whereby the government claimed that it had the support of the people, when it did not. The first relates to the introduction of direct elections into the Legislative Council in 1988, and the second relates to the introduction of subversive laws in 2003. In both cases, the government conducted open consultation, and over 100,000 submissions were received by the government each time, involving some 230,000 to 370,000 individual signatures. In both cases, the government engineered the result of the analysis and lied with statistics.


Although 16 years apart, these two events tell more or less the same story. Without the check and balance by professional bodies specialized in opinion polling, and without the final "verification" of the public wish by popular votes, there is always the temptation for government to engineer public opinion in its so-called "consultation" process.




Hong Kong was a British colony up to June 30, 1997, thereafter it became a "Special Administrative Region" (SAR) under the People's Republic of China. Under the Joint Declaration signed by Britain and China in 1984, which entered in force on May 27, 1985, the HKSAR would be placed under "one country, two systems", and would "enjoy a high degree of autonomy".


Back in 1984, major democratic elections in Hong Kong only took place at the "District Board" level, and as explained, such elections were only introduced in 1982. District Boards in Hong Kong were consultative bodies which advised the government on local municipal affairs. They were renamed as "District Councils" after the handover in 1997. There are now 18 District Councils in the whole of Hong Kong.


Britain and China began their negotiations on the future of Hong Kong in 1982, and not long after the opening of such talks, the British Hong Kong Government embarked on a comprehensive review of Hong Kong's constitutional development. In July 1984, it published a consultative document entitled Green Paper: The Further Development of Representative Government in Hong Kong. Paragraph 6 of the Green Paper said,


"The purpose of this Green Paper is to suggest how the central institutions of government in Hong Kong might be made more representative in a way which will make the Government more directly accountable to the people of Hong KongˇK"


In November 1984, one month before the Sino-British agreement was officially signed, the Hong Kong Government concluded its consultation by publishing its final decisions in a White Paper. It basically decided to expand the Legislative Council in 1985 to include 57 members, of whom 12 were elected by an electoral college, 12 by functional constituencies, 22 unofficial members appointed by the Governor, 10 official members, and the Governor as President. The White Paper also promised to review the progress of constitutional development in 1987.


Meanwhile, public awareness and interest grew in constitutional reform, and a number of pressure groups emerged to demand for greater and faster reform. China, however, was skeptical of such reforms and decided to take a hard line against them. As one political scientist in Hong Kong puts it, the Beijing Government was clearly apprehensive about Britain's intentions in the remaining years of colonial rule, and may have thought that the Green and White Papers of 1984 were British strategies to establish a more autonomous government, not necessarily pro-Chinese, before 1997 (Scott, 1989: 285-6, in So, 1999: 126).


It was in this context that the Hong Kong Government issued a Green Paper in May 1987 entitled The 1987 Review of Developments in Representative Government. By then, the government has lost the will to press ahead for comprehensive constitutional reform, yet it had to honour its words in 1984 that a review would take place in 1987 as scheduled. Its main strategy was to tune down its enthusiasm for rapid reforms by changing the tone and format of its consultation exercise. In the 1987 Green Paper, it simply listed a large number of options for introducing changes into the top legislature, and the major controversial issue of introducing direct elections in 1988 was only placed as a minor item in one of the many options. As Scott (1989: 293) puts it, the government tried to bury the issue "in a welter of minor, non-controversial constitutional options and proposals."


But what is most relevant to this study is that the Green Paper also promised to consult the general public on all the options, and that the government would determine the pace of reform in accordance with the wish of the community. To facilitate its collection and collating of public opinion, a Survey Office was established to receive written submissions from the public over a four-month period. It turns out that the Survey Office also commissioned a market research company to conduct two opinion polls on the proposals. The strategies used by the government in manipulating these opinion polls will be discussed in the next section of this paper.


We will be brief on what happened next. In 1988, the government finally decided to delay the introduction of direct elections for three years, to 1991. As Tsang (1995: 95) observes, the government's emphasis "was changed from promoting greater democracy to maintaining continuity beyond 1997." From then to 1997, electoral politics gradually developed in Hong Kong, but not without controversy. After Chris Patten became the last Governor of Hong Kong in 1992, he again tried to speed up democratic development by expanding the franchise of functional constituencies to become quasi-popular elections. Such a move displeased the Chinese Government, and immediately after the handover of 1997, the Legislative Council was disbanded. The first HKSAR Legislative Council was voted into existence in 1998, using a different electoral system. It had a brief term of two years, and elections were held again in 2000. The next general election is due to be held in September 2004. In January 2003, about sixteen years after the 1988 Direct Election Saga, another incident of similar nature occurred - the "Article 23 Saga".


The Basic Law is a sort of mini-constitution for the HKSARG. It was promulgated in April 1990, and came into effect on July 1, 1997, after the handover of sovereignty. Article 23 of the Basic Law stipulates that the HKSAR "shall enact laws on its own to prohibit any act of treason, secession, sedition, subversion against the Central People's Government, or theft of state secrets, to prohibit foreign political organizations or bodies from conducting political activities in the Region, and to prohibit political organizations or bodies of the Region from establishing ties with foreign political organizations or bodies." It didn't stipulate when and how.


Because of the sensitivity nature of the issue, it soon became apparent after the handover that legislation for Article 23 would not occur during the first term of office of HKSAR's Chief Executive, up to mid-2002.


However, shortly after the Chief Executive started his second term of office, a decision was made that legislation should be passed within a year, that is, in 2002-2003. Government officials started to lobby in the summer of 2002 that they would publish a consultation document soon but gave assurance that they would take a minimalist approach and no one needed to worry.


On September 24, 2002, the government published a consultation document entitled Proposals to Implement Article 23 of the Basic Law and asked the people to send in their views and comments by December 24, 2002. The document raised many concerns. At the later stage of the consultation period, many people began to ask for a White Bill so that they could see how the law would be drafted. The consultation drew 100,909 submissions (of which 97,097 were local), representing views of 369,612 individuals (340,513 from Hong Kong) and over 1,000 organizations. On January 28, 2003, the government published The Compendium of Submissions on Article 23 of the Basic Law, and claimed that majority of the public was in favour of the legislation. It also refused to publish a White Bill, claiming that most respondents did not ask for it.


On February 26, 2003, the government introduced the National Security (Legislative Provision) Bill to the Legislative Council, and expected it to be passed in early July the latest. Even though there was a severe outbreak of SARS (a form of atypical pneumonia) in Hong Kong between March and May, the government refused to delay legislation. Public opposition began to mount.


On July 1, about 500,000 Hong Kong people marched on the street to protest against the hasty passage of the Bill. About a week later, on July 5, the Chief Executive announced that the government would give three major concessions but insisted on the passage of the Bill on July 9. Less than 48 hours later, leader of the Hong Kong Liberal Party, James Tien, resigned from the Executive Council, and declared that his party would not support the Bill. Without their critical eight votes in the Legislative Council, the Chief Executive announced hours later that the Bill would be withdrawn.


The force of July 1 continued to roam, and two more large scale protests were held. On July 16, the government official primarily responsible for introducing the Article 23 legislation, Secretary for Security Mrs Regina Ip, resigned. Another senior official, Financial Secretary Antony Leung, who was involved in a financial scandal in March, also resigned hours later. The two persons, together with the Chief Executive, were the major targets of people's demonstration.


On September 5, the Chief Executive finally announced his decision to withdraw the Bill from the legislative programme, "so as to allow sufficient time for the community to study the enactment question". He also said that the Security Bureau would establish a special working group to review afresh the legislative work.




The 1988 Direct Election Saga started with the publication of The 1987 Review of Developments in Representative Government by the Hong Kong Government on May 27, 1987. According to the official yearbook published by the Hong Kong Government, "The purpose of the review was to consider whether the systems of representative government in Hong Kong should be developed in 1988 and, if so, in what manner" (Hong Kong 1988: Ch. 2). To demonstrate the importance of the consultation exercise, the government made an unprecedented move to set up a "Survey Office" for the collection and organization of public opinion. It also happened that the Survey Office also commissioned a commercial market research firm, AGB McNair Hong Kong Ltd, to conduct two large-scale public opinion surveys. By the end of the consultation period, a total of 131,589 submissions were received, and on November 4, 1987, the Survey Office published its report.


On the crucial question of whether or not to introduce direct election into the Legislative Council in 1988, the Survey Office reported that 94,270 individuals were opposed to it, while 39,345 were in favour. The "no" apparently had it, with a majority of 7:3. This however, was a result of deliberate engineering, as the government adopted two major strategies to rig the results, in order to please China. The first strategy was to collate the submissions in a way which downplayed the strength of opinion in favour of direct election, the second strategy was to use a loaded questionnaire in its public opinion surveys to compress favourable response for the introduction of direct elections.


Put it simple, the Survey Office classified the opinions gathered into three categories, namely, "submissions from organizations and individuals", "opinion surveys", and "petitions". It then concluded that the opposition view was dominant among submissions under the first two categories, while the 21 petitions, involving 23,866 supporters and 295 opponents, were said to be of very little value because they were products of mobilization. Opinions gathered from public opinion surveys were said to be more useful, and based on two biased surveys commissioned by the Survey Office itself, it concluded that people surveyed did not have a clear preference. This contradicted all opinion surveys conducted by community groups at that time, which found that people supported introducing direct elections in 1988. On the other hand, for the category of "submissions from organizations and individuals", the Survey Office grouped 73,767 pre-printed standard letters with more or less identical opinion under this category, and concluded that the opposition view was dominant. However, massive submissions on pre-printed standard letters are by nature not much different from "petitions", they ought not have been classified under this category under the government's own logic.


Had the government given equal weight to the standard letters and signature forms by classifying them impartially under the third category, category one and three would have become dominated by submissions in support of direct elections. What's more, had the government conducted its opinion polls according to prevailing professional standard, "submissions" under category two would also have been in favour of introducing direct elections. In other words, in terms of broad categories, a 3-0 absolute victory for direct election advocates was engineered to become a 1-2 marginal defeat.


To grasp the complexity of the issue, and hence the government's "wisdom" in the move, we can look at the figures presented by Legislative Councillor Chan Ying-lun at the Legislative Council debate on the Survey Office's report on November 18, 1987. Chan made a "re-analysis" of the figures to debunk the government strategy, as follows:


Re-analysis of the information contained in Paragraphs 13.26-13.43 of the Survey Office report:

  Support for direct election in 1988 Objection to direct election in 1988 Rates (approx. no. after decimal point) Remarks
 Legislative Council
 Urban Council
 Regional Council
 District boards
Total ratio
 No. of submissions from groups and associations 229 352 3.9:6.1  
 Total no. of submissions from individuals and groups of individuals 35,129 84,202 2.9:7.1 Categorisation being challenged thus further classify as follows
 No. of submissions (excluding the preprinted and identical) from individuals and groups of individuals 33,816
(35,129 -1,313)
(84,202 -66,269)
6.5:3.5 The no. of preprinted identical submissions is deducted from the total
 No. of submissions (preprinted and identical) from individuals and groups of individuals 1,313 66,269 0.2:9.8 The ratio only reflects the view of the preprinted and identical submissions
 Signature collected from signature campaigns 223,866 295 9.9:0.1  
 8 random sampling surveys conducted by commercial research companies 41%-62% 16%-27%    
 Survey conducted by concerned staff of Tertiary Education Institutions for Constitutional Development 19% 33%   Organising body disagree with the classification of the Survey Office
 25 surveys covering specific population groups 16%-81% 6%-45%    
 124 non-random sampling surveys covering individuals 26,529 15,401 6.3:3.7  
 4 random sampling surveys covering groups, associations and other bodies 431 479 4.7:5.3  
 AGB McNair HK Ltd 1st Survey 15% 19% 4.4:5.6  
 AGB McNair HK Ltd 2nd Survey 12% 21% 3.6:6.4  

It can be seen that if "pre-printed and identical" submissions were taken out from "submissions from organizations and individuals", the ratio between "for" and "against" would be reversed from 2.9 : 7.1 to become 6.5 : 3.5 . While 98% of these pre-printed letters were against direct election, but if they were analyzed in conjunction with submissions from signature campaigns (since they were all mobilized), the "supporters" would have a clear majority of 7.7 : 2.3 .


During the same debate, another Legislative Councillor Martin Lee presented the results of 9 territory-wide probability sampling surveys on the question of direct elections in 1988, to confront the AGB McNair results. The "don't knows" were left out in his analysis, as follows:


Opinion surveys conducted by the community:

  Sponsor* Pollster* For Against
 (1) CSTEI CSTEI 38% 62%
 (2) SCMP FSA 58% 42%
 (3) SCMP MDR 54% 46%
 (4) SCMP MDR 66% 34%
 (5) SCMP MDR 69% 31%
 (6) 4 papers SRH 77% 23%
 (7) 4 papers SRH 76% 24%
 (8) 4 papers SRH 68% 32%
      65% 35%
 (9) 4 papers SRH 69% 31%
      68% 32%

* The exact names could be found in the Legislative Council proceedings. They are immaterial to Martin Lee's argument, and are not given here.


Martin Lee then explained that the figures presented against CSTEI survey should not be taken to indicate respondents' objection to introducing direct elections in 1988, because respondents could choose from 1987, 1988, 1989 and all the way until 1997, and the figures showed that 1988 was the most favoured year. He explained that it would be wrong to group all other percentages for the other years together as the total percentages against the introduction of direct elections in 1988. The important point made by Martin Lee was that practically all opinion polls commissioned by community groups during that period showed a marked difference from the result of the government polls.


Notwithstanding such challenges, the government still concluded that it was not the wish of the Hong Kong people to introduce direct elections to the Legislative Council in 1988. The situation was eloquently summarized by Chris Patten, the last Governor of Hong Kong, in his memoir:


"The results of the 'consultation process' that took place were officially interpreted to mean that Hong Kongers did not want direct elections in 1988, despite the fact that this is what a majority of them appeared to favour in the government's own survey and in every independent survey carried out at the same time. The statistical contortions required to arrive at this outcome, which satisfied both the British and Chinese governments, have been admirably and regularly exposed over the years." (1998: 36)


As to why the government rigged the results, Patten explained:


"Before the consultation process was completed, the Chinese believed they had a secret deal with Britain: if it was concluded at the end of this public debate that direct elections were not to be held in 1988 then the Chinese would include a commitment to direct elections in their own Basic Law. They were sufficiently receptive to British sensitivities to keep this under wraps. For their part, the British did not believe they had made an actual deal, but they did think there was an understanding that if Hong Kong opinion appeared not to favour direct elections in 1998 then the Chinese would make the Basic Law commitment. No deal, so far as Britain was concerned: just an understanding between friends - a secret understanding." (1998: 36)




Probably because of the lesson learned in 1988, no "Survey Office" was set up again in Hong Kong in all subsequent consultation exercises. In 2002-03, the consultation of proposals to implement Article 23 of the Basic Law was undertaken completely by the Security Bureau. Contrary to the usual practice, the HKSAR Government did not commission any opinion survey, independent or otherwise, to measure people's support of the proposals. Instead, the HKSAR Government simply published a consultation document on September 24, 2002 entitled Proposals to Implement Article 23 of the Basic Law and asked the people to send in their views and comments by December 24, 2002. By the end of that period, a total of 100,909 submissions (of which 97,097 were local), representing views of 369,612 individuals (340,513 from Hong Kong) and over 1,000 organizations, were received and collated in the Compendium of Submissions on Consultative Document on "Proposals to Implement Article 23 of the Basic Law". In between, the supporting and opposing camps alike staged extensive mobilization campaigns, and tens of thousands of people took to the streets at one point or another.


Because the government claimed that it was important "to give free rein to the public to express their views" (Compendium Volume 1, Page 1), the public was not consulted in any structured way, and the trick again lied in the classification of the over 100,000 submissions collected. This time, the government classified the submissions according to their opinion into three types, and according to their format into four categories. In terms of content, those supporting legislating on Article 23 or proposals in the consultation document were classified as Category A; those opposing legislating or the proposals as Category B; and those without explicit views supporting or opposing legislating or the proposals as Category C. In terms of format, the four categories were: (1) submissions from organizations, (2) submissions from individuals, (3) submissions in the form of standard letters or pre-printed opinion forms, and (4) signature forms.


Using 100,909 submissions as the base, the government calculated that 67.5% of them belonged to Category A, 28.2% to B, and 4.3% to C, and concluded that the general public was in favour of the proposals. However, if we use the number of signatures embedded in those submissions as the base, then, of the 369,612 signatures received, 36.9% belonged to A, 60.2% to B, and 2.8% to C, and the "no" should have it. Although the Compendium did list all the figures, just like the British Hong Kong Government in 1987, the HKSAR Government down-played petitions collected via "signature forms", up-played standard pre-print standard letters, and engineered the conclusion at its own wish.


As on whether a White Bill should be issued to further consult the public, the government produced these figure to support its conclusion that no White Bill was necessary -


For White Bill - (i) Individuals 15.1% (ii) Organizations 5.6% (iii) Standard letters/printed forms 6.2% and (iv) Signature forms 2.8%.


Against White Bill - (i) Individuals 6.8% (ii) Organizations 10.8% (iii) Standard letters/printed forms 6.9% and (iv) Signature forms 0.5%.


Preference not indicated - (i) Individuals 78.1% (ii) Organizations 83.6% (iii) Standard letters/printed forms 86.9% and (iv) Signature forms 96.4%.


Immediately after the publication of the Compendium, there was uproar, two academics at the University of Hong Kong quickly raised the following questions (Chung and Choy, January 29, 2003) -


1) Why did the government not use the overall number of signatures collected in deriving its conclusions?


2) Why should there be a distinction between "pre-printed forms" and "signature forms"?


3) Why, in all media sessions, did the government only referred to local submissions, and neglected non-local submissions with 29,099 signatures predominantly against the proposals?


4) Submissions were classified as Category A if they "supported legislating on Article 23 or proposals in the consultation document", and Category B if they "oppose legislating or the proposals". What if someone supports legislation but opposes the proposals? Or supports legislation in principle, but opposes kicking-off the process right now? Would that submission be classified as Category A or B?


5) According to the government, the majority of Hong Kong people did not have a view on the issue of a White Bill. But if someone opposed legislation, was there a need for this person to express a view on the White Bill issue?


After some months, a group of volunteer academics who called themselves "The Research Team on the Compendium of Submissions on Article 23 of the Basic Law" re-analyzed all the submissions, and contrasted the government statistics with their own, as follows (from the Report of the Research Team, June 10, 2003):


Comparison of views reported in compendium, views on the principle of legislation and views on the content of consultation document for local submissions:

  Organization Independent letter Standard letter Signature form
  Submission % Submission % Signature % Signature % Signature %
 Support reported 940 84.6 2,857 56.7 5,173 57.2 66,609* 77.5* 65,185* 26.5*
 Oppose reported 83 7.5 1,374 27.3 1,859 20.6 16,332* 19.0* 175,823* 71.4*
 Uncertain reported 87 7.8 804 15.9 2,006 22.2 3,046* 3.5* 5,124* 2.1*
 Total 1,110   5,035   9,038   85,987*   246,132*  
 Support principle 989 89.0 3,117 61.8 5,454 60.3 53,490 65.9 67,693 26.6
 Oppose principle 93 8.4 1,407 27.9 2,388 26.4 12,034 14.8 182,789 71.8
 Uncertain principle 29 2.6 518 10.3 1,204 13.3 15,649 19.3 4,031 1.6
 Support content 942 84.8 2,864 56.8 5,169 57.1 53,461 65.9 67,633 26.3
 Oppose content 151 13.6 1,998 39.6 3,672 40.6 27,679 34.1 183,962 72.5
 Uncertain content 18 1.6 180 3.6 205 2.3 33 0.0 2,918 1.1
 Total 1,111   5,042   9,046   81,173   254,513  

* Based on counts and percentages reported in the Compendium compiled by the government.


Comparison of views reported in compendium, views on the principle of legislation and views on the content of consultation document for overseas submissions:

  Organization Independent letter Standard letter Signature form
  Submission % Submission % Signature % Signature % Signature %
 Support reported 1 1.0 20 1.6 50 2.9 0* 0.0* 0* 0.0*
 Oppose reported 78 79.6 1,041 85.3 1,420 81.8 1,671* 100.0* 25,327* 98.6*
 Uncertain reported 19 19.4 157 12.9 263 15.2 0* 0.0* 288* 1.4*
 Total 98   1,218   1,733   1,671*   25,615*  
 Support principle 2 2.0 26 2.1 53 3.1 4 0.4 91 0.5
 Oppose principle 87 88.8 1,140 93.7 1,626 93.7 951 95.3 19,726 99.5
 Uncertain principle 9 9.2 51 4.2 52 3.0 43 4.3 0 0.0
 Support content 1 1.0 17 1.4 44 2.50 4 0.4 0 0.0
 Oppose content 95 96.9 1,179 96.6 1,666 96.00 994 99.6 19,817 100.0
 Uncertain content 2 2.0 11 0.9 11 0.60 0 0.0 0 0.0
 Total 98   1,217   1,731   998   19,817  

* Based on counts and percentages reported in the Compendium compiled by the government.


During the re-analysis, the government's two-way classification was carefully scrutinized. All submissions were re-read and re-classified according to whether they supported or opposed Article 23 legislation "in principle", and whether they supported or opposed to the specific proposals given in the government consultative document. The first table reports the figures for local submissions, and the second for overseas submissions.


It can be seen that while the government's classification was by an large correct, there was only majority support for the government proposals among local organizations' submissions (around 85%), local standard letters (around 66%), and local independent letters (around 57%). All forms of submissions from non-local respondents were against the proposals (97% to 100%), and petitions collected through signature forms were overwhelmingly against the proposals (around 73% for locals and 100% for overseas). If we merge all submissions together and count the signatures, using the "one man one vote" concept, then roughly 60% were against the government proposals, and only about 37% supported them. By claiming that Hong Kong people were supportive of the proposals, the government has once again lied with statistics.




The two incidents reported in this paper took place sixteen years apart, but they were extremely similar. They demonstrate how an undemocratic government could manipulate opinion statistics to show that they have the people's support, when in fact it does not.


In 1987, the British Hong Kong Government took the trouble to conduct large scale opinion surveys, with loaded questionnaires, to demonstrate that they had the support of the mass. In 2003, the HKSAR Government did not bother to do so, probably because they knew that, unlike in 1987, independent academics and pollsters would keep a watchful eye on all government opinion surveys. However, as with the British Hong Kong Government before them, the SAR Government has resorted to classify public submissions at will, and twisted the wish of the community by mis-classification submissions and mis-interpreting statistical figures.


To tackle these problems, the Research Team on the Compendium of Submissions on Article 23 of the Basic Law of which the author was a member, made the following comments and suggestions in their Report,


"The major problem of the government's consultation has been the failure to use an appropriate method to collect quality opinions. Consultation documents usually do not provide options nor discussion points concerning the specific legislation proposals for public discussion, resulting in citizens' oversimplified expression of their stands and emotionalizing the discussion process and further enhancing social polarization."


"During consultation, the government did not clarify how the opinions would be analyzed. Different ways of counting can result in different interpretation. This allowed the government to cheat in the process, and damaged the credibility of consultations."


In particular, the Research Team made the following suggestions for large-scale policy consultations in future:


1) The government should state the key questions clearly and explicitly, preferably providing clear options as well.


2) The government should state its method of analysis before conducting the consultation, in order to be fair in treating all opinions expressed.


3) The government should appoint an independent consultant or organization to collect and collate public opinion, in order to enhance the credibility of the process.


Needless to say, shall any opinion poll be conducted to measure the public's reception of the policy proposals, be they conducted by the government of community groups, international standards like those endorsed by WAPOR should be adhered to, in order to prevent opinion polls being misused or abused.


These measures, however, may not be able to overcome the problem in the long run. At the very end, a government not returned by popular election, hence without people's mandate, is always tempted to manipulate opinion figures to order to demonstrate that it indeed has the mandate of the people de facto. If such claims cannot be put to test by open and fair elections, the government will always try to rig the results of public consultation, or findings of opinion polls, in its favour. If they could not be done easily, the government might ban the polls altogether. In open but undemocratic societies like Hong Kong, it is vitally important for the media and pollsters to introduce the international standards to their societies, and then adhere to them.




Chan, Jennifer So-Kuen (2003). "Let Figures Speak For Themselves - Reanalysis on the Public Opinion Towards Article 23 Consultation Document", available from, June 6, 2003. Hong Kong.


Chan, Kin-Man and Chung, Robert Ting-Yiu (2003). "Redress for the Public Voice", available from, June 11, 2003. Hong Kong.


Choy, Boris Sai-Tsang and Chung, Robert Ting-Yiu (2003). "Casting Doubts on the Compendium of Submissions", available from, January 29, 2003. Hong Kong.


Hong Kong Government and HKSAR Government. Hong Kong Annual Reports, various Green Papers, White Papers, and Consultative Documents. Hong Kong Government Press. Hong Kong.


Hong Kong Legislative Council. Official Report of Proceedings (1987-88 Session). Hong Kong.


Patten, Chris (1998). East and West: The last Governor of Hong Kong on Power, Freedom and the Future. MacMillan. London.


Research Team on the Compendium of Submissions on Article 23 of the Basic Law (2003). Doing Justice to Public Opinion in Public Consultations: What to Do and What NOT to Do - A Case Study of the Government's Consultation Exercise on its Proposals to Implement Article 23 of the Basic Law (Enhanced version). June 10, 2003. Hong Kong.


So, Alvin Y. (1999). Hong Kong's Embattled Democracy: A Societal Analysis. Johns Hopkins University Press. Baltimore.


Scott, Ian (1989). Political Change and the Crisis of Legitimacy in Hong Kong. Oxford University Press. Hong Kong.


Schumpeter, Joseph R. (1950). Capitalism, Socialism, and Democracy, third edition. (First edition 1942.) Harper & Row. New York.


Spangenberg, Frits (2003). The Freedom to Publish Opinion Poll Results: Report on a Worldwide Update. ESOMAR (The World Association of Research Professionals) & WAPOR (World Association for Public Opinion Research).


Tsang, Steve (ed., 1995). Government and Politics: A Documentary History of Hong Kong. Hong Kong University Press. Hong Kong.


Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady (1995). Voice and Equality: Civic Voluntarism in American Politics. Harvard University Press. Cambridge, Mass.




Robert Ting-Yiu Chung is the Director of Public Opinion Programme at the University of Hong Kong, Hong Kong, he is also the WAPOR Representative from Hong Kong. Mailing address: Public Opinion Programme, the University of Hong Kong, Pokfulam Road, Hong Kong.
E-mail address: [email protected]. Webpage:

Everything carried in this website does not represent the stand of the University of Hong Kong. Dr Robert Ting-Yiu Chung, Director of Public Opinion Programme, is responsible for everything posted herewith, except for column articles and messages posted at the opinion platforms. Any responsibility, legal and otherwise, arising from any such item shall vest entirely with its author. The intellectual property of all information contained in this website, including questionnaire design and research reports, originally belongs to HKU POP unless specified otherwise. HKU POP opens their copyright to the world via this website, but would request all users to acknowledge their source properly.

HKU POP all rights reserved. This page is maintained by [email protected]. Last update :  30/12/2011