| Background | Data Analysis | Commentary |
Background
Since its establishment in 1991, the Public Opinion Programme (POP) at the University of Hong Kong has been conducting different types of opinion studies on social and political issues, as well as providing research services for different organizations, on condition that POP would design and conduct all studies independently, and could also release the findings for public consumption. In November 2009, POP came into agreement with the "now News Channel" for a project called "Joint Public Opinion Research Project on Political Reform". The main objective of the project is to demonstrate, by focusing on ongoing discussions of political reform, how independent research institute and professional news media investigate, analyze, report and comment on public opinion, including the explanation and promotion of professional ethics of opinion studies. The project comprises conducting regular and ad-hoc opinion surveys and other public opinion studies after the launching of public consultation by the government. The survey results will first be released in the "now News Channel", followed by POP press releases for public consumption. The "now News Channel" agrees to POP uploading these programmes to the POP Site for public education, while POP agrees that "now News Channel" uses these findings for productions without POP's involvement.
Since December 7, 2009, the results of the project are released once every Monday. Today POP releases the results for the fourth time, but this time on its own, which mainly focus on the analysis of collective resignation in 5 constituencies and functional constituencies with respect to the respondents' social strata. POP believes that the analysis can help people better understand the diversity and complexity of recent public opinion, as part of its public education effort.
|
Data Analysis
Our social strata analysis has used data coming from three different surveys, herewith the contact information of each survey:
Date of survey |
Sample base |
Overall response rate |
Sampling error of percentages* |
19-23/11/09 |
1,001 |
76.7% |
+/-3% |
28/11-4/12/09 |
1,017 |
74.5% |
+/-3% |
14-17/12/09 |
1,000 |
70.5% |
+/-3% |
* "95% confidence stratum" 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.
In each of the above surveys, we asked the respondents to self-define which one of the five social classes they belong to. The options were: upper class, upper middle class, mid-middle class, lower middle class and lower class or grassroots. According to the choice of respondents, we further classified them into relatively upper stratum, relatively middle stratum and relatively lower stratum, or upper stratum, middle stratum and lower stratum in short. Each stratum contains about one-third of the sample. The following table shows the choice of respondents mapped to their relative social stratum:
Social class self-reported by respondents |
Relative social stratum used in our analysis |
Upper class |
Relatively upper stratum (about one-third) |
Upper middle class |
Mid-middle class |
Lower middle class |
Relatively middle stratum (about one-third) |
Lower class or grassroots |
Relatively lower stratum (about one-third) |
Don't know/hard to say/refuse to answer |
Not classified |
Collective resignations in 5 LegCo constituencies
Date of survey: 28/11-4/12/09 |
Upper stratum |
Middle stratum |
Lower stratum |
Overall sample |
Q1. Support or oppose LC members resign to induce a de facto referendum? |
Support |
27% +/-5%
(93) |
25% +/-5%
(76) |
26% +/-5%
(88) |
26% +/-3%
(259) |
Half-half |
9% +/-3%
(33) |
8% +/-3%
(25) |
10% +/-3%
(35) |
9% +/-2%
(93) |
Oppose |
55% +/-5%
(190) |
55% +/-6%
(167) |
45% +/-5%
(152) |
51% +/-3%
(516) |
Don't know/ hard to say |
9% +/-3%
(32) |
12% +/-4%
(36) |
18% +/-4%
(61) |
14% +/-2%
(142) |
Total |
100%
(348) |
100%
(304) |
100%
(336) |
100%
(1,010) |
Date of survey: 28/11-4/12/09 |
Upper stratum |
Middle stratum |
Lower stratum |
Overall sample |
Q2. Will elect the resigned democratic LC member or his/her recommended candidate to return to LC? |
Yes |
37% +/-5%
(128) |
38% +/-6%
(116) |
34% +/-5%
(113) |
36% +/-3%
(364) |
Half-half |
3% +/-2%
(11) |
5% +/-2%
(14) |
4% +/-2%
(13) |
4% +/-1%
(38) |
No |
53% +/-5%
(184) |
47% +/-6%
(142) |
46% +/-5%
(153) |
49% +/-3%
(491) |
Don't know/ hard to say |
7% +/-3%
(25) |
10% +/-3%
(31) |
17% +/-4%
(57) |
12% +/-2%
(117) |
Total |
100%
(348) |
100%
(304) |
100%
(336) |
100%
(1,010) |
Date of survey: 14-17/12/09 |
Upper stratum |
Middle stratum |
Lower stratum |
Overall sample |
Q3. The Democratic Party has decided not to resign collectively, should other democratic parties continue to resign collectively? |
Yes |
24% +/-4%
(88) |
30% +/-5%
(93) |
28% +/-5%
(80) |
27% +/-3%
(263) |
No |
55% +/-5%
(201) |
52% +/-6%
(162) |
45% +/-6%
(130) |
51% +/-3%
(504) |
Don't know/ hard to say |
21% +/-4%
(78) |
19% +/-4%
(58) |
27% +/-5%
(79) |
23% +/-3%
(224) |
Total |
100%
(367) |
100%
(313) |
100%
(289) |
100%
(991) |
Analysis shows that in our survey completed in early December, a respective of 55%, 55% and 45% of people from upper, middle and lower stratum opposed the collective resignation of 5 constituencies of the democratic camp, and 53%, 47% and 46% of people from upper, middle and lower stratum respectively said they would not vote for the democratic candidates in the by-elections. For the survey in mid-December conducted after the Democratic Party had decided not to support the collective resignation in 5 constituencies, 55%, 52% and 45% of people from the upper, middle and lower stratum respectively regarded other democratic parties should not collectively resign.
The issue of retention or abolition of functional constituencies
Date of survey: 14-17/12/09 |
Upper stratum |
Middle stratum |
Lower stratum |
Overall sample |
Q4. Discuss the issue of retention or abolition of functional constituencies before voting for the political reform proposal, or discuss after passing it? |
Discuss the issue after passing the political reform proposal |
40% +/-5%
(146) |
28% +/-5%
(89) |
30% +/-5%
(86) |
33% +/-3%
(326) |
Discuss the issue before voting for the political reform proposal |
44% +/-5%
(161) |
55% +/-6%
(173) |
44% +/-6%
(128) |
48% +/-3%
(472) |
Don't know/ hard to say |
16% +/-4%
(60) |
16% +/-4%
(51) |
26% +/-5%
(75) |
20% +/-3%
(193) |
Total |
100%
(367) |
100%
(313) |
100%
(289) |
100%
(991) |
For the issue of retention or abolition of functional constituencies, analysis shows that, 40% of people from the upper stratum said we should discuss the issue after passing the political reform proposal, 55% from the middle stratum thought we should discuss it before voting for the political reform proposal, while 26% of people from the lower stratum said they had no idea.
|
| Commentary
Note: The following commentary is written by the Director of Public Opinion Programme Dr Robert Chung in simple Q&A format:
Q: Why analyze the data using social strata?
A: The reason for such analyses is to detect any rift which exists between different social strata. From the angle of policy formulation, such analysis can show the degree of resistance from specific social groups on different policies, as well as the degree of support from different groups on specific political actions.
Q: Is it realistic to separate society into upper, middle and lower stratum?
A: It is a profound academic question of how to divide society into different strata. It is suggested that the society should be classified into capitalists and proletariats, or to further classify the capitalists into high, middle, and petite bourgeoisie, or into upper, upper-middle, middle, lower-middle, and working classes. In POP's standard questionnaires, we ask the respondents to self-define one of the 5 social classes they belong to, namely, upper, upper middle, mid-middle, lower middle and lower class or grassroots respectively. After considering the size and actual meaning of each sub-sample, we further classify them into upper, middle and lower social stratum according to their self-reported answers. Each stratum contains about one-third of the total sample.
Q: On collective resignation in 5 constituencies proposed by the democrats, what is the orientation of people from each stratum? Are there changes after the Democratic Party made its decision?
A: Our first survey shows that people in the upper and middle strata have the strongest opposition against collective resignation, with 55% in each stratum, and more people in the lower stratum answered "don't know". After the Democratic Party made its decision, our second survey shows that people in the middle stratum changed their views most, as some of them changed from agreeing with the opinion of upper stratum to agreeing with the lower stratum. That means the number of people in the middle stratum supporting collective resignation has increased, whereas people in the other strata have changed little.
Q: If the democratic camp resigns in 5 constituencies, which stratum will more likely vote for democratic candidates in the by-elections?
A: Our result shows that about 30% of people in each stratum said they would vote for the democrats. However, 53% in the upper stratum said they would definitely not vote for them, which is the highest figure across three strata. Relatively less people in the middle and lower strata said the same, while the lower stratum has the most stray votes, at 17%, which is higher than the 10% of the middle stratum and 7% of the upper stratum.
Q: As for the issue of retention or abolition of the functional constituencies, what is the orientation of people from each stratum?
A: Based on our cross-tabulation analyses, most people in the middle stratum want to discuss the issue of retention and abolition of the functional constituencies as soon as possible, at 55%. Relatively more people in the upper stratum want to defer the discussion, and relatively more people in the lower stratum had no idea, with 40% and 26% respectively. |
| Background | Data Analysis | Commentary |
| |
