"Legislative Council Election Exit Poll Analysis Article Series" No.2Back

Hong Kong Island Exit Poll Results

Robert Ting-Yiu Chung and Carmen Ka-Man Chan
(Members of Public Opinion Programme, the University of Hong Kong)
Translated by Carmen Ka-Man Chan
(Research Executive, Public Opinion Programme, the University of Hong Kong)

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


In the first article of this series, we analyzed the different demands and characteristics of voters who supported the democratic camp, the pro-China camp and the Liberal Party, as part of our quest for a new consensus. From this article onward, we will concentrate on the specific circumstances within each geographical constituency, especially on the characteristics of voters in support of different candidate lists. We hope our analyses would provide a scientific and comprehensive basis for us to understand the election results. We will start with Hong Kong Island first.


The concerned poll was conducted on the election day. Our research team has successfully interviewed 9,223 people who have just voted, together with 998 partially completed cases, in 101 polling stations in Hong Kong. The overall response rate was 66.7%. Among them, 1,800 successful cases and 252 partially completed cases were extracted from 20 polling stations on Hong Kong Island, with a response rate of 67.1%. All figures quoted in this article came from the group of more than 2000 cases above. We will not go into the research design here as all information of the survey will be published in our HKU POP Site


It must be stated that, just like the first article of this series, no. 2 Rita Fan has been listed as belonging to the "pro-China camp" while no. 5 Wong Kam-Fai has not been classified. Besides, since the support ratings of both no. 3 Tsang Kin-Shing and no. 5 Wong Kam-Fai were less than 2%, this article will not analyze the characteristics of their source of votes. This should not affect much.


After the election, the result of Hong Kong Island immediately aroused public attention. Strategy of tactical voting failed in the democratic camp and this led to the defeat of Cyd Ho despite her high support rating obtained. Yeung Sum, Chairman of the Democratic Party (DP), became target of a thousand arrows. Some criticisms were made out of good intentions, whereas some people just criticized to instigate and sow dissension. As a neutral researcher, we have no intention to appreciate or criticize the election engineering or electoral strategy of any parties. Also, we will not express appreciation or pity towards the election result. We just want to examine and test the "truth" generally believed by the pubic through objective figures.


Regarding the question of tactical voting, our exit polls have recorded some very important figures. Table 1 shows that, among the supporters of Yeung Sum, 20% claimed vote allocation was their main consideration. This was a very high percentage. Among candidate lists in all constituencies, this was only lower than the 31% obtained by Bruce Liu in Kowloon West, while the respective percentages of other lists were at most around 10%. Yet, Bruce Liu only obtained 13,452 votes and thus only around 4,000 votes were involved in tactical voting. Basing on the 131,788 votes obtained by Yeung Sum's list in the end, 26,000 votes were obtained from tactical voting. Assuming voters of the democratic camp allocated their votes only between Yeung Sum's and Audrey Eu's lists, while 4%, i.e. 3, 000 votes were also obtained from tactical voting in Audrey Eu's list, the net value of votes obtained from tactical voting in Yeung Sum's list should thus be 23,000. If the votes could be shared more evenly, or the 5,315 votes obtained by Tsang Kin-Shing's list could all be allocated to the democratic camp, then the election result would be completely changed. Of course, if we assume the democratic camp could re-allocate the votes, we should also assume the pro-China camp could re-allocate the votes. If both two camps could accomplish this perfectly, then the last seat of Hong Kong Island would hinge on 3,000 to 5,000 votes, and the democratic camp would gain the upper hand.


If one wants to examine the tactical voting behaviour of the voters, apart from directly asking the voters their main considerations, we could also get some hints by analyzing the background information of voters. Table 2 shows that, in the pro-China camp's lists on Hong Kong Island, 64% of supporters of Ma Lik's list were male, while 58% of Rita Fan's supporters were female. In the democratic camp's lists, among the supporters of Yeung Sum's list, 60% were male, while 56% of supporters of Audrey Eu's list were female. Before the election, the democratic camp had once advocated "Male vote for male, female vote for female". This vote allocating strategy was seemingly a half success. If DP had not made strong calls to appeal for votes at the final stage, the concerned result would be more apparent. However, what is interesting is that, although the pro-China camp had not advocated the same method, its supporters seemed to have allocated their votes according to gender also, and this was even more apparent than the democratic camp.


Looking back at Table 1, among the supporters of Ma Lik's list, 53% said their main consideration was political party factor, while the corresponding percentages in Yeung Sum's list, Audrey Eu's list and Rita Fan were 38%, 6% and 5%. In other words, the party sense among supporters of Ma Lik's list was the strongest, Yeung Sum's list followed, whereas Audrey Eu's list and Rita Fan's supporters mainly considered the candidates' quality. If we combined the "political party factor" (38%) and "vote allocation consideration" (20%) of Yeung Sum's list, then the party sense of their supporters was not weak as well. On Hong Kong Island, DP and Democratic Alliance for the Betterment of Hong Kong (DAB) seemed to be capable to influence around 60% of their supporters in their names. This could be viewed as the threshold of their vote allocation.


Between political and economic-livelihood issues, Table 3 reveals that, DAB and DP were apparently in two extremes of a political spectrum. A respective of 28% and 66% of their supporters were politics-oriented. The corresponding percentages of Rita Fan and Audrey Eu's list were 38% and 52%. The former was closer to DAB while the latter was closer to DP. Such spectrum could also be found in other political indicators, including the proportion of voters taking part in the July 1 Demonstration, the demands towards universal suffrage, as well as the appraisal towards the HKSAR Government.


An analysis on demographic factors shows that more first-time voters could be found among the supporters of Rita Fan and Audrey Eu's list. Overall speaking, more young voters (18-29 years old) supported Audrey Eu's list, more middle-aged voters (30-59 years old) supported Yeung Sum's list and more aged voters (60 years old or above) supported Ma Lik's list. Regarding the voters' academic attainment, in general, supporters of Ma Lik's list had a lower education level whereas supporters of Audrey Eu's and Yeung Sum's lists possessed a higher education level, while supporters of Rita Fan lay in between. Concerning the voters' occupation, more executives and professionals supported Yeung Sum's list, more manual workers supported Ma Lik's list while students were more likely to support Audrey Eu's list. However, if not for the final vote allocation, the background of voters in support of Yeung Sum's list and Audrey Eu's list might be more similar.


The last important analysis is to examine the two major camps' flow of votes over the last 4 years. Table 4 selects the lists which have won in the last election and compares them with the voters' choices in this election. Figures reveal that, 7% of votes obtained by DP in the last Legislative Council election have flowed to the pro-China camp this year, while 12% of votes obtained by DAB have flowed to the democratic camp. The source of votes of both sides was rather stable. For voters supporting Cyd Ho in the last election, excluding 15% which have flowed to Rita Fan, around half of the remaining 85% have flowed to DP, which was seemingly a support of the appeal of vote allocation. However, only 19% of DP's votes have flowed to Audrey Eu's list. Probably this explained why the democratic camp's vote allocation strategy failed on Hong Kong Island.


The exit poll reveals that, 30% of voters have not voted in the last Legislative Council election, but have voted in this year's. Regarding these voters' orientation, the ratio of the democratic camp's to the pro-China camp's supporters was around 6:4, which was coherent to the overall orientation of the voters. To put it in another way, the new voters seemingly have not affected the balance between the two camps. For the situation in other constituencies, we will leave them in the coming articles.


Table 1: Main considerations for choosing candidates

    Ma Lik's list   Rita Fan   Yeung Sum' s list   Audrey Eu's list 
  Candidate factor  20%   89%   26%   73% 
  Political party factor  53%   5%   38%   6% 
  Combination of lists  10%   2%   12%   15% 
  Vote allocation concern  8%   2%   20%   4% 
  Hard to say  10%   3%   3%   2% 

Table 2: Voters' background - Proportion of:

    Ma Lik's list   Rita Fan   Yeung Sum' s list   Audrey Eu's list 
  Male  64%   42%   60%   44% 
  Female  36%   58%   40%   56% 
  Voters possessing tertiary education level or above  29%   45%   53%   58% 
  Executives and professionals  31%   36%   52%   43% 
  Manual workers  10%   4%   4%   4% 
  Full-time housewives  4%   12%   6%   14% 
  Voters who have not voted before  25%   38%   26%   33% 


Table 3: Voters' political orientation - Proportion of voters who:

    Ma Lik's list   Rita Fan   Yeung Sum' s list   Audrey Eu's list 
  Chose basing on candidates' political attitude  28%   38%   66%   52% 
  Chose basing on candidates' economic and livelihood policies  60%   48%   26%   41% 
  Took part in July 1 Demonstration  11%   14%   54%   41% 
  Supported direct election of CE in 2007  16%   24%   65%   43% 
  Supported direct election of CE in 2012  31%   39%   16%   27% 
  Supported direct election of all LC members in 2008  22%   35%   69%   50% 
  Supported direct election of all LC members in 2012  25%   24%   10%   17% 
  Were satisfied with HKSARG  32%   16%   2%   2% 
  Were "half" satisfied with HKSARG  52%   55%   23%   40% 
  Were dissatisfied with HKSARG  13%   28%   75%   56% 

Table 4: Flow of votes in 2 elections

    Martin Lee's list   Gary Cheng's list   Cyd Ho   Have not voted last time 
  Ma Lik's list  3%   68%   0%   17% 
  Rita Fan  4%   19%   15%   23% 
  Yeung Sum's list  72%   5%   40%   31% 
  Audrey Eu's list  19%   7%   43%   26%