Back

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

The author, together with the colleagues of the Public Opinion Programme, has written a series of 6 articles on the analysis of the Legislative Council election and in fact we can stop at this point. Yet, since there are still some people in the society who have misunderstood our work, and even confused our survey with the election engineering of the political parties, the author thus would like to have a discussion on a few basic questions concerning the exit polls as a conclusion of this series of articles.

Except for information concerning the HKU Public Opinion Programme and the Student Research Team, all other information quoted in this article were extracted from open information. The author had not inquired about any news from any political parties or candidates. What concerns the author most is the professional development of the public opinion polls.

Before the introduction of the Legislative Council geographical direct election in 1991, whether the media should be allowed to report the exit poll during the whole election day had soon become the focus of argument between the government and the electronic media. As the representative of the research agency, the author also participated in some top-level meetings at that time. At that time, the government had no legal basis to ban exit polls, but individual officials had urged the citizens not to respond to those exit polls. Thanks to the media bodies at that time, the freedom of information was preserved in Hong Kong.

After the election then, the author had written an article to discuss the experiences from the exit polls in the United States and examine their effect to the electoral behaviour of the voters. As there is a time lag between the Eastern and the Western coasts in the United States, when the exit poll result is released in the Eastern coast, voting is still going on in the Western coast. However, as the freedom of speech is granted by the constitution, the Federal Court still vetoes any measures which will limit the conduct of exit polls and the release of the their results. Nevertheless, due to the media's professionalism and self-control, they will not release the exit polls' forecast of that state before the close of voting in individual states.

After many years' practice, a consensus has been formed gradually between the media bodies and the government in Hong Kong. The government will not legislate to ban the release of the exit poll result for the whole day, while the media bodies will not anticipate the result before the close of election as well. If the media fails to observe the regulations, the Electoral Affairs Commission (EAC) can condemn it publicly. This practice, which seeks to resort to the moral and public power instead of the legal ban, is the same as the system in the United States. We should treasure this.

According to the guidelines issued by the EAC, for organizations intended to conduct exit polls, they only need to submit the personal information of the person-in-charge and the interviewers, as well as the selected stations, to the EAC 7 days before the election, then they can conduct exit polls around the polling stations on the election day, while activities beyond the "No Canvassing Zone" are not limited. This year, the EAC has raised its transparency and released all the names of organizations and workers involved which have applied for conducting exit polls. With the help of these additional public information, the author could analyze some rarely noticed figures here.

According to the EAC's information, apart from the HKU Public Opinion Programme and the HKU Student Research Team, which are headed and supported by the author respectively, 5 more organizations conducted exit pill on the election day. They were the Hong Kong Policy Research Institute, Tsuen Wan Youth Council, Hong Kong Social and Economic Research Centre, Hong Kong Youths Unified Association, and Public Affairs Research Society. Before the District Council election last year, the author has exchanged ideas with the person responsible for exit polls in the Hong Kong Policy Research Institute. For the other organizations, the author has not got in touch with them.

Basing on the information on the website of the EAC, together with the internal information of the 2 research teams in HKU, we come up with a frequency table on human resources. Since there are quite a lot of overlaps in the EAC's information, the author could only be as accurate as possible. The table shows that, among the 7 research teams, the Public Affairs Research Society had the largest manpower. The Hong Kong Social and Economic Research Centre, and then Hong Kong Youths Unified Association followed. Nearly 2,000 people had been sent out from these 3 organizations to over 300 stations to conduct exit polls. When one takes a careful look at the selected stations of these 3 organizations, their district distribution was clear-cut. Even though 2 organizations were found in Kowloon East at the same time, there was completely no overlap of selected stations. The author had not investigated the background of these organizations, but their resources deployed were really surprising.

The exit polls conducted by the HKU Public Opinion Programme over these 13 years aimed at meeting the needs of both the academic research and media coverage. Therefore, the questionnaire we used was relatively complex and long, and the result would be fully published. We have all along advocated a scientific sampling method, so as to minimize the disturbance to the respondents and the other research agencies. Take this election as an example, we have only sent 107 workers and conducted interviews in 101 polling stations. The human resource involved was only one-twentieth of the 3 main organizations'.

To maintain absolute neutrality, the author has never released the result to the candidates before the close of the election. Even the media bodies which have sponsored the exit polls would only know the preliminary forecast of the election result at around 9:00 pm. The sponsors have also promised not to release the result before the close of the election. In view of the candidates' appeal to the voters basing on the exit poll result after half of the election day these years, the HKU Public Opinion Programme deliberately issued press release on September 10 this year to state that all similar slogans were irrelevant to us.

However, after the election, since the democratic camp experienced a failure in the vote allocation on Hong Kong Island, the media has kept on suspecting whether we have informed the Democratic Party the exit poll result. One column article said, "At 7:00 pm the reporter phoned Law Chi-Kwong to inform him the news. Even though he thought this was the fake news released by the DAB, he should be curious to know Martin Lee's situation. Why didn't he make a call to ask Robert Chung in HKU who was conducting the exit poll…"

Later, Martin Lee of the Democratic Party openly explained in this column article that the Democratic Party did not conduct exit polls, and the critical situation of Cyd Ho was only known from the DAB's exit poll result released by his friends working in the media. On the other side, on the day after the election, Yip Kwok-Him of the DAB publicly admitted that the DAB has conducted exit polls and has known from the exit poll result that Martin Lee was greatly ahead of Choy So-Yuk at about 4:30 pm on the election day. The public speeches of the 2 people have eventually done justice to the author.

The author has never objected any political parties or other agencies to conduct exit polls. On the contrary, being able to base policy platforms and election engineering on scientific figures is the advancement of the society. What the author opposes, is the research agencies' use of dishonest means to gauge voters' opinions for their election engineering secretly. As a matter of fact, the author would like to share his experiences with any agencies, whether think-tanks or political organizations, which are interested in conducting these researches. One should know, vicious competition will only lead to waste of resources and annoyance to voters, and even ruin the fairness of the election system.

The consensus reached between the EAC and the media bodies all along these years is that there will be no forecast of result before the close of the election. Although some media academics have suspected whether this kind of arrangement will limit the freedom of information, this kind of "Hong Kong model" works well so far. However, if individual candidates are able to obtain some valuable information through some channels, or one side or both sides keep on using the exit poll result to support their own appeals and vote allocation strategies, the instructions set by the EAC to prohibit the release of exit poll result will become meaningless then?

Some time ago, there were always noises and crowds outside the polling stations in Hong Kong. Some people believe that this kind of polling activity can create the atmosphere and raise people's intention to vote. Nowadays, this kind of thought has become out-dated. Voters generally want to enter the polling stations silently to cast their sacred votes. At present, some people advocate setting a "cool-down" period for the election, aiming at establishing a set of peaceful and rational electoral culture so as to be in line with the advanced society.

Among the democratic countries which the author knows, the political parties will rarely conduct exit polls on the election day to match their last-minute election engineering. If a society's election result completely hinges on the voters' emotional response in the last stage (such as a shooting case), or the last 12 hours' election engineering (such as candidates' appeals to voters with their critical situation), that will be a great misfortune.

Hopefully our society can seek self-improvement and get rid of those unhealthy activities which are detrimental to the professionalism and unhelpful to the growth of civic wisdom.

Frequency Table: Human Resources Analysis of Organizations Conducting Exit Polls

		HKU Public Opinion Programme	HKU Student Research Team	HK Policy Research Institute	Tsuen Wan Youth Council	HK Social and Economic Research Centre	HK Youths Unified Association	Public Affairs Research Society
HKI	No. of selected stations	20	7	5	0	33	0	0
	No. of fixed interviewers	20	14	12	0	375	0	0
	No. of floating interviewers	6 (the whole HK)	9 (the whole HK)	0	0	73 (on HKI)	0	0
KLNE	No. of selected stations	16	5	6	0	7	13	0
	No. of fixed interviewers	16	11	13	0	21	60	0
	No. of floating interviewers	6 (the whole HK)	9 (the whole HK)	0	0	0	7 (in Kln)	0
KLNW	No. of selected stations	13	4	2	0	0	14	0
	No. of fixed interviewers	13	8	4	0	0	100	0
	No. of floating interviewers	6 (the whole HK)	9 (the whole HK)	0	0	0	7 (in Kln)	0
NTE	No. of selected stations	24	10	13	0	0	0	118
	No. of fixed interviewers	24	18	30	0	0	0	545
	No. of floating interviewers	6 (the whole HK)	9 (the whole HK)	0	0	0	0	0
NTW	No. of selected stations	28	5	4	9	0	0	121
	No. of fixed interviewers	28	11	9	29	0	0	741
	No. of floating interviewers	6 (the whole HK)	9 (the whole HK)	0	0	0	0	0
Total	No. of selected stations	101	31	30	9	40	27	239
	No. of workers	107	71	68	29	469	167	1,286