2010 Legislative Council By-election Survey First Summary ReportBack

 
Press Release on April 26, 2010

| Background | Survey Method | Contact Information | Propensity to vote |
| Candidate strength in each constituency | Commentary |


Background

To fulfill its social responsibility, the Public Opinion Programme (POP) at the University of Hong Kong began to conduct its 2010 Legilative Council by-election survey on April 8, due to end on May 12. Because of resources constraint, the survey is conducted like an irregular quasi rolling poll with small samples size per day but covering a long fieldwork period. When sufficient data is accumulated, the findings will be released in the form of summary reports. All findings will be uploaded onto the "HKU POP SITE" (http://hkupop.pori.hk) for public consumption. The summary report released today covers the survey period of April 8 to 25.


Survey Method

Like other telephone surveys conducted by POP, these LC by-election surveys are conducted by real interviewers under close supervision. Target population of these surveys are general population of all the geographical constituencies aged 18 or above. Telephone numbers are first selected randomly from telephone directories as seed numbers. Another set of numbers is then generated by using the plus/minus one/two method. Duplicate numbers are then screened out, and the remaining numbers mixed in a random order to become the final sample. Upon successful contact being made with a target household, one member of the household is selected among those present using the "next birthday" rule.

As a general practice, all figures have been weighted according to provisional figures obtained from the Census and Statistics Department regarding the gender-age distribution of the Hong Kong population in 2009 year-end. All the result findings are based on the weighted figures.


Contact Information

POP interviewed 1,612 Hong Kong citizens including 1,277 registered voters about their views on 2010 Legislative Council by-election between April 8, 9, 12, 15-25 by means of random telephone interviews conducted by real interviewers. The sampling error of figures using the overall sample is below +/-3% at 95% confidence level, and the response rate of the survey is 67%. Contact information of the survey is as follows:

Date of survey

Overall sample size

Response rate

Maximum sampling error of percentages*

8-25/4/10

1,612(including 1,277
registered voters)

67.1%

+/-3%

* Errors are calculated at 95% confidence level using full sample size. "95% confidence level" means that if we were to repeat a certain survey 100 times, using the same questions each time but with different random samples, we would expect 95 times getting a figure within the error margins specified.


Propensity to vote

The survey reveals that registered voters' propensity to vote is 43%, which is a significant drop from that registered at a similar period before the 2008 LC election. However, this percentage should not be taken as an indication of the actual turnout rate. A propensity of 43% would not give a 43% turnout rate, because many people who claim they would vote at this stage would eventually not vote.

Territory-wide sample

Date of survey

10-18/8/08

8-25/4/10

Difference

Sample base

3,269

1,277**

--

Response rate

70.9%

67.1%

--

Sampling error of percentages*

+/- 2%

+/- 3%

--

Proportion of respondents definitely will vote

57% +/-2%

25% +/-2%

-32%#

Propensity to vote (summation of "definitely will" and "most likely will" vote)

84% +/-1%

43% +/-3%

-41%#

Constituency-specific samples

HKI

KLW

KLE

NTW

NTE

Response rate

238

190

193

335

321

Sampling error of percentages*

+/- 6%

+/- 7%

+/- 7%

+/- 5%

+/- 6%

Proportion of respondents definitely will vote

25% +/-6%

27% +/-6%

27% +/-6%

24% +/-5%

25% +/-5%

Propensity to vote (summation of "definitely will" and "most likely will" vote)

43% +/-6%

42% +/-7%

45% +/-7%

43% +/-5%

44% +/-6%

* All error figures in the table are calculated at 95% confidence level. "95% confidence level" means that if we were to repeat a certain survey 100 times, using the same questions each time but with different random samples, we would expect 95 times getting a figure within the error margins specified.
** sub-sample base of registered voters.
# Such changes have gone beyond the sampling errors at the 95% confidence level, meaning that they are statistically significant prima facie. However, whether numerical differences are statistically significant or not is not the same as whether they are practically useful or meaningful.



Candidate strength in each constituency

In this analysis, "firm votes" is defined as voters who said they would definitely or very likely vote, and have already decided how to vote, while "stray votes" is defined as voters who said they would definitely or very likely vote, but have not decided how they would vote. "Excludable votes" is defined as those who did not fall into either of these categories. Statistics for the five constituencies are as follows:

Hong Kong Island

   Date of survey:

10-18/8/08

Error margin*

8-25/4/10

Error margin*

Difference*

   Sample base:

671

--

238

--

--

   Sampling error of percentages*

+/- 4%

--

+/- 6%

--

--

   Proportion of firm votes

66%

+/-4%

31%

+/-6%

-35%#

   Proportion of stray votes

22%

+/-3%

12%

+/-4%

-10%#

   Proportion of excludable votes

12%

+/-2%

57%

+/-6%

+45%#

Kowloon West

   Date of survey:

10-18/8/08

Error margin*

8-25/4/10

Error margin*

Difference*

   Sample base:

525

--

190

--

--

   Sampling error of percentages*

+/- 4%

--

+/- 7%

--

--

   Proportion of firm votes

60%

+/-4%

27%

+/-6%

-33%#

   Proportion of stray votes

23%

+/-4%

15%

+/-5%

-8%#

   Proportion of excludable votes

17%

+/-3%

58%

+/-7%

+41%#

Kowloon East

   Date of survey:

10-18/8/08

Error margin*

8-25/4/10

Error margin*

Difference*

   Sample base:

605

--

193

--

--

   Sampling error of percentages*

+/- 4%

--

+/- 7%

--

--

   Proportion of firm votes

62%

+/-4%

37%

+/-7%

-25%#

   Proportion of stray votes

21%

+/-3%

8%

+/-4%

-13%#

   Proportion of excludable votes

17%

+/-3%

55%

+/-7%

+38%#

New Territories West

   Date of survey:

10-18/8/08

Error margin*

8-25/4/10

Error margin*

Difference*

   Sample base:

778

--

335

--

--

   Sampling error of percentages*

+/- 4%

--

+/- 5%

--

--

   Proportion of firm votes

61%

+/-4%

31%

+/-5%

-30%#

   Proportion of stray votes

22%

+/-3%

11%

+/-3%

-11%#

   Proportion of excludable votes

18%

+/-3%

57%

+/-5%

+39%#

New Territories East

   Date of survey:

10-18/8/08

Error margin*

8-25/4/10

Error margin*

Difference*

   Sample base:

690

--

321

--

--

   Sampling error of percentages*

+/- 4%

--

+/- 6%

--

--

   Proportion of firm votes

63%

+/-4%

32%

+/-5%

-31%#

   Proportion of stray votes

19%

+/-3%

12%

+/-4%

-7%#

   Proportion of excludable votes

18%

+/-3%

56%

+/-6%

+38%#

* All error figures in the table are calculated at 95% confidence level. "95% confidence level" means that if we were to repeat a certain survey 100 times, using the same questions each time but with different random samples, we would expect 95 times getting a figure within the error margins specified.
# Such changes have gone beyond the sampling errors at the 95% confidence level, meaning that they are statistically significant prima facie. However, whether numerical differences are statistically significant or not is not the same as whether they are practically useful or meaningful.


Statistics for the strength of candidates in each constituency are as follows:

Question: In the forthcoming Legislative Council by-election, there are several candidates as follows. Which one do you tend to support? (Interviewers to READ OUT all candidates' lists)

Raw support rate

Error margin*

Effective support rate

Error margin*

Hong Kong Island #

   Date of survey

8-25/4/10

   Sample base

233

   Sampling error of percentages*

+/- 6%

   Tanya Chan

42%

+/-6%

91%

+/-5%

   Tai Cheuk-yin

1%

+/-2%

3%

+/-3%

   Lee Chun-hung

1%

+/-1%

2%

+/-3%

   Leung Wing-ho

1%

+/-1%

2%

+/-3%

   Wong Hing

1%

+/-1%

2%

+/-3%

   Not yet decide

19%

+/-5%

 

 

   Others^

36%

+/-6%

 

 

Kowloon West

   Date of survey

8-25/4/10

   Sample base

184

   Sampling error of percentages*

+/- 11%

   Wong Yuk-man

27%

+/-7%

61%

+/-11%

   Peck Wan-kam

11%

+/-5%

26%

+/-10%

   Lam Yi-lai

2%

+/-2%

5%

+/-5%

   Shea Kai-chuen

1%

+/-1%

2%

+/-3%

   Kwok Shiu-ming

1%

+/-1%

2%

+/-3%

   Chiang Sai-cheong

1%

+/-1%

1%

+/-2%

   Wong Weng-chi

1%

+/-1%

1%

+/-3%

   Cheung Kam-hung

0%

+/-1%

1%

+/-2%

   Not yet decide

22%

+/-6%

 

 

   Others^

34%

+/-7%

 

 

Kowloon East #

   Date of survey

8-25/4/10

   Sample base

188

   Sampling error of percentages*

+/- 7%

   Leong Kah-kit Alan

47%

+/-7%

89%

+/-6%

   Lai King-fai

6%

+/-3%

11%

+/-6%

   Not yet decide

13%

+/-5%

 

 

   Others^

34%

+/-7%

 

 

New Territories West

   Date of survey

8-25/4/10

   Sample base

330

   Sampling error of percentages*

+/- 6%

   Chan Wai-yip

37%

+/-5%

82%

6%

   Li Kwai-fong

5%

+/-2%

11%

5%

   Chow Ping-tim

1%

+/-1%

3%

3%

   Li Sai-hung

1%

+/-1%

3%

3%

   Kwok Wing-kin

1%

+/-1%

1%

2%

   Not yet decide

19%

+/-4%

 

 

   Others^

36%

+/-5%

 

 

New Territories East

   Date of survey

8-25/4/10

   Sample base

317

   Sampling error of percentages*

+/- 8%

   Leung Kwok Hung

31%

+/-5%

69%

+/-8%

   Chan Kwok-keung

8%

+/-3%

18%

+/-7%

   Wu Sai-chuen

3%

+/-2%

7%

+/-4%

   Chow Ching Crystal

2%

+/-2%

6%

+/-4%

   Not yet decide

16%

+/-4%

 

 

   Others^

40%

+/-6%

 

 

* All error figures in the table are calculated at 95% confidence level. "95% confidence level" means that if we were to repeat a certain survey 100 times, using the same questions each time but with different random samples, we would expect 95 times getting a figure within the error margins specified.
^ Included responses of "won't vote" and "won't support any political parties or candidates".
# Chan Yuet-tung and To Sum were candidates of HK Island and Kowloon East constituencies respectively in the first place. Since they have been verified as invalid by Election Affairs Office, they are disgarded from this table.



Commentary

Note: The following commentary was written by Director of POP Robert Chung.

To fulfill its social responsibility, POP today releases its survey findings on the forthcoming Legislative Council by-election. Due to resources constraint, we are not running not a daily rolling poll but an irregular quasi rolling poll with small sample size per day but covering a long fieldwork period starting from the close of nomination on April 8. When sufficient data is accumulated, we will release our findings in the form of summary reports. Our first summary report released today covers the period of April 8, 9, 12, and 15 to 25, concurring with the first half of the campaign period. A total of 1,612 valid samples has been collected.

Our survey shows that at the first half of the campaign period, 43% of all registered voters intend to vote, with little variation across constituencies. If we would depreciate the figure by half or one-third to calculate the final turnout rate, we would have a figure between 22% and 29%, which is much lower than that of 2008.

Regarding candidate strengths, we first filter registered voters in our sample, and then locate the "firm", "stray" and "excludable votes" in each constituency. "Firm votes" is defined as voters who said they would definitely or very likely vote, and have already decided how to vote; "stray votes" is defined as voters who said they would definitely or very likely vote, but have not decided how to vote, while "excludable votes" is defined as those who do not fall into either of these categories. In our survey, "excludable votes" stands high at 55% to 60% across different constituencies, which is rare.

From our analysis, although our sample size is small and sampling errors high, because voting propensity is low, candidate strengths are quite lopsided across all constituencies. The situation is a bit tighter in Kowloon West and NT East, because Wong Yuk-man and Leung Kwok-hung are only leading Peck Wan-kam and Chan Kwok-keung by 16 and 23 percentage points respectively in their raw support rates, with sampling errors standing at 7 and 5 percentage points. There are still three more weeks before the election, whether voters' propensity to vote would change, and hence alters the trends of "firm" and "stray" votes remains to be seen.

| Background | Survey Method | Contact Information | Propensity to vote |
| Candidate strength in each constituency | Commentary |