HKU POP releases findings of Budget instant pollBack
Press Release on March 1, 2018
| Detailed Findings (Budget Feature Page) |
Abstract
The Public Opinion Programme (POP) at The University of Hong Kong interviewed 614 Hong Kong people yesterday (February 28), including 252 landline samples and 76 mobile samples by means of a random telephone survey conducted by real interviewers, plus 286 panel samples by online survey. Some of its findings have already been released via our “HKU POP SITE” (http://hkupop.pori.hk) last night, today we release the more comprehensive and in-depth analyses. Our latest Budget instant survey shows that, among those who had some knowledge of the second Budget Speech by FS Paul Chan, 26% said they were satisfied, 41% said they were not, giving a net satisfaction rate of negative 14 percentage points, which goes down significantly by 29 percentage points from his first Budget last year. On a scale of 0-100, this Budget scored 48.2 marks, which is also significantly worse than that of last year, down by 7.5 marks. The rating of the instant survey shows that people’s appraisal of this Budget has reached the record low since the survey question was first asked in 2008. As for Paul Chan’s own popularity after the Budget, both his support rating and net approval rate have not changed much since early February, meaning that the Budget has not much instant impact on his popularity. Further analyses show that the younger the respondents, the less satisfied they are with the Budget and less supportive towards Paul Chan. Our instant survey describes people’s instant reaction towards the Budget, how people’s reaction will change after knowing more about the Budget will be revealed by our next follow-up survey. The maximum sampling error of all percentages is +/-4 percentage points at 95% confidence level, while the sampling errors of rating figures and net values need another calculation. The response rate of the survey is 68%.
Points to note:
[1] The address of the “HKU POP SITE” is http://hkupop.pori.hk, journalists can check out the details of the survey there.
[2] The sample size of this survey is 614 successful interviews, not 614 x 67.5% response rate. In the past, many media made this mistake.
[3] The maximum sampling error of all percentages is +/-4 percentage points at 95% confidence level, while the sampling error of rating figure and net value needs another calculation. “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. When quoting these figures, journalists can state “sampling error of rating not more than +/-2.3, sampling error of percentages not more than +/-4%, and sampling error of net values not more than +/-7% at 95% confidence level”.
[4] Because of sampling errors in conducting the survey(s) and the rounding procedures in processing the data, the figures cannot be too precise, and the totals may not be completely accurate. Therefore, when quoting percentages of the survey(s), journalists should refrain from reporting decimal places, but when quoting the rating figures, one decimal place can be used.
[5] The data of this survey is collected by means of random telephone interviews conducted by real interviewers, not by any interactive voice system (IVS). If a research organization uses “computerized random telephone survey” to camouflage its IVS operation, it should be considered unprofessional.
Background
Since 1992, POP has been conducting Policy Address instant surveys every year. From 1998 onwards, we expanded our instant surveys to cover the Budget Speeches. In free and democratic societies, instant surveys are indispensable sources of free information. Combined with appropriate follow-up surveys, and in parallel to expert analyses, they give a multi-dimensional picture of opinion development. They are an important part of a society’s interactive development. Starting 2008, we split up previous years’ instant survey into two surveys. In our first survey, we measure people’s overall appraisal of the Budget, their rating of the Budget, their change in confidence towards Hong Kong’s future, and FS’s popularity. In our second survey, we focus on people’s reactions towards different government proposals, their satisfaction with the government’s fiscal policies, and other relevant issues. Starting 2011, we revised our design to concentrate on people’s appraisal of the Budget and FS’s popularity in our instant survey, and move the remaining questions to our follow-up survey. There is no change to our operation this year.
Latest Figures
POP today releases via the “POP SITE” the findings of the Budget instant poll. From July 2017, POP enhanced the previous weighting method that has been used for quite a few years. Apart from age, gender and education, economic activity group is now also taken into account when adjusting data. The latest figures released today have been rim-weighted according to figures collected in the 2016 By-census regarding the gender-age distribution of the Hong Kong population and the 2017 educational attainment (highest level attended) distribution and economic activity status distribution obtained from the Census and Statistics Department. The mobile sample and the panel sample have also been rim-weighted according to the basic Public Sentiment Index (PSI) figures collected in the landline sample. Herewith the contact information for the latest survey:
|
Date of survey |
Effective sample size |
Effective response rate |
Maximum sampling error
|
|
28/2/2018 |
614 |
67.5% |
+/-4% |
[6] 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. Questions using only sub-samples would have bigger sampling error. Sampling errors of ratings are calculated according to the distribution of the scores collected.
People’s satisfaction figures with this year’s Budget are summarized below together with the previous findings:
|
Date of survey |
Sub-sample
|
Appraisal of Budget[7] |
|||||
|
Satisfaction rate[9] |
Half-half[9] |
Dissatisfaction
|
Net satisfaction rate |
Mean value[9] |
Satisfaction rating of Budget |
||
|
28/2/18 |
551 |
26[10]+/-4% |
28+/-4% |
41[10]+/-4% |
-14[10]+/-7% |
2.7[10]+/-0.1 (Base=523) |
48.2[10]+/-2.2 |
|
22/2/17 |
502 |
33+/-4% |
30+/-4% |
18+/-3% |
15+/-6% |
3.2+/-0.1 (Base=412) |
55.7+/-2.1 |
|
24/2/16 |
500 |
36[10]+/-4% |
31+/-4% |
20+/-4% |
17[10]+/-7% |
3.2+/-0.1 (Base=432) |
57.2[10]+/-2.1 |
|
25/2/15 |
529 |
45[10]+/-4% |
28+/-4% |
18[10]+/-3% |
28[10]+/-6% |
3.3[10]+/-0.1(Base=483) |
60.2[10]+/-1.8 |
|
26/2/14 |
695 |
24[10]+/-3% |
26[10]+/-3% |
45[10]+/-4% |
-20[10]+/-6% |
2.7[10]+/-0.1(Base=660) |
49.8[10]+/-1.7 |
|
27/2/13 |
813 |
30[10]+/-3% |
37[10]+/-3% |
31[10]+/-3% |
-1[10]+/-5% |
2.9[10]+/-0.1 (Base=793) |
53.6[10]+/-1.5 |
|
1/2/12 |
826 |
38[10]+/-3% |
33+/-3% |
26[10]+/-3% |
12[10]+/-6% |
3.1[10]+/-0.1 (Base=804) |
57.0[10]+/-1.4 |
|
23/2/11 |
911 |
27[10]+/-3% |
34+/-3% |
35[10]+/-3% |
-8[10]+/-5% |
2.8[10]+/-0.1 (Base=884) |
51.5[10]+/-1.5 |
|
24/2/10 |
724 |
47[10]+/-4% |
35[10]+/-4% |
14[10]+/-3% |
32[10]+/-5% |
3.4[10]+/-0.1 (Base=698) |
60.8[10]+/-1.4 |
|
25/2/09 |
669 |
30[10]+/-4% |
43[10]+/-4% |
22[10]+/-3% |
8[10]+/-6% |
3.1[10]+/-0.1 (Base=636) |
54.8[10]+/-1.5 |
|
27/2/08 |
811 |
68[10]+/-3% |
21[10]+/-3% |
5[10]+/-2% |
63[10]+/-4% |
3.8[10]+/-0.1 (Base=768) |
70.6 +/-1.2 |
|
28/2/07 |
673 |
62[10]+/-4% |
25 +/-3% |
9[10]+/-2% |
53[10]+/-5% |
3.6[10]+/-0.1 (Base=649) |
-- |
|
22/2/06 |
577 |
50 +/-4% |
26 +/-4% |
19[10]+/-3% |
31+/-6% |
3.3+/-0.1 (Base=550) |
-- |
|
16/3/05 |
544 |
47[10]+/-4% |
29 +/-4% |
11 +/-3% |
36[10]+/-6% |
3.4+/-0.1 (Base=478) |
-- |
|
10/3/04 |
395 |
37[10]+/-5% |
33[10]+/-5% |
12[10]+/-3% |
25[10]+/-7% |
3.3[10]+/-0.1 (Base=324) |
-- |
|
5/3/03 |
495 |
20[10]+/-4% |
23 +/-4% |
50[10]+/-4% |
-30[10]+/-7% |
2.5[10]+/-0.1 (Base=459) |
-- |
|
6/3/02 |
539 |
47[10]+/-4% |
23 +/-4% |
17 +/-3% |
30[10]+/-6% |
3.3[10]+/-0.1 (Base=472) |
-- |
|
7-8/3/01 |
263 |
57[10]+/-6% |
25[10]+/-5% |
13[10]+/-4% |
44[10]+/-9% |
3.5[10]+/-0.1 (Base=248) |
-- |
|
8/3/00 |
643 |
70[10]+/-4% |
12[10]+/-3% |
4[10]+/-2% |
66[10]+/-4% |
3.9[10]+/-0.1 (Base=553) |
-- |
|
3/3/99 |
598 |
46[10]+/-4% |
27[10]+/-4% |
10[10]+/-2% |
36[10]+/-5% |
3.4[10]+/-0.1 (Base=496) |
-- |
|
18/2/98 |
638 |
55 +/-4% |
20 +/-3% |
7 +/-2% |
47+/-5% |
3.6+/-0.1 (Base=520) |
-- |
[7] 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.
[8] Excluding respondents who did not answer this question because they had not heard of / did not have any knowledge of the Budget. Sampling errors of ratings are calculated according to the distribution of the scores collected. Because of the smaller sample size, the sampling error has increased accordingly.
[9] Collapsed from a 5-point scale. The mean value is calculated by quantifying all individual responses into 1, 2, 3, 4, 5 marks according to their degree of positive level, where 1 is the lowest and 5 the highest, and then calculate the sample mean.
[10] 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 is not the same as whether they are practically useful or meaningful, and different weighting methods could have been applied in different surveys.
After excluding those respondents who said they did not have any knowledge of this year’s Budget, this year’s instant survey showed that 26% were satisfied with it, 41% were dissatisfied and 28% said “half-half”, giving a net satisfaction rate of negative 14 percentage points. The mean score is 2.7, which is between “half-half” and “quite dissatisfied” in general. Meanwhile, the average rating registered for the Budget was 48.2 marks.
Figures on various Financial Secretaries’ popularity before and after their Budget Speeches since 2001 are summarized as follows: [11]
|
Popularity of Donald Tsang |
Popularity of
|
Popularity of Henry Tang |
|||||||||||||||
|
Year of the Budget |
2001-02 |
2002-03 |
2003-04 |
2004-05 |
2005-06 |
2006-07 |
2007-08 |
||||||||||
|
Date of Budget Speech[12] |
7/3/2001 |
6/3/2002 |
5/3/2003 |
10/3/2004 |
16/3/2005 |
22/2/2006 |
28/2/2007 |
||||||||||
|
FS’s rating at Budget instant survey & error [13] |
69.7+/-1.6 |
63.4+/-1.0 |
49.8+/-1.2 |
59.9+/-0.8 |
63.3+/-1.0 |
63.0+/-1.2 |
64.1+/-1.4 |
||||||||||
|
FS’s net approval rate at Budget instant survey & error [13] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
59+/-4% |
56+/-5% |
56+/-5% |
||||||||||
|
Date of the latest survey before Budget Speech [14] |
19-21/2/2001 |
18-21/2/2002 |
1-4/3/2003 |
1-3/3/2004 |
1-3/3/2005 |
3-7/2/2006 |
1-6/2/2007 |
||||||||||
|
FS’s rating before the Budget & error [13] |
71.9+/-0.9 |
57.2+/-1.2 |
48.1+/-1.3 |
57.4+/-1.0 |
59.7+/-1.0 |
63.0+/-1.0 |
60.8+/-1.2 |
||||||||||
|
FS’s net approval rate before the Budget & error [13] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
57+/-4% |
50+/-4% |
||||||||||
|
Change in
|
-2.2 [15] |
+6.2 [15] |
+1.7 [15] |
+2.5 [15] |
+3.6 [15] |
-- |
+3.3 [15] |
||||||||||
|
Change in FS’s
|
[Not applicable] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
[Not applicable] |
-1% |
+6% [15] |
||||||||||
|
Popularity of John Tsang |
|||||||||||||||||
|
Year of the Budget |
2008-09 |
2009-10 |
2010-11 |
2011-12 |
2012-13 |
2013-14 |
2014-15 |
2015-16 |
2016-17 |
||||||||
|
Date of Budget Speech[12] |
27/2/2008 |
25/2/2009 |
24/2/2010 |
23/2/2011 |
1/2/2012 |
27/2/2013 |
26/2/2014 |
25/2/2015 |
24/2/2016 |
||||||||
|
FS’s rating at Budget instant survey & error [13] |
67.9+/-1.5 |
54.9+/-1.6 |
61.3+/-1.4 |
52.4+/-1.3 |
54.1+/-1.2 |
56.6+/-1.3 |
54.0+/-1.4 |
61.0+/-1.7 |
62.2+/-1.8 |
||||||||
|
FS’s net approval rate at Budget instant survey & error [13] |
54+/-5% |
28+/-6% |
46+/-5% |
13+/-5% |
3+/-5% |
35+/-4% |
27+/-6% |
44+/-6% |
48+/-6% |
||||||||
|
Date of the latest survey before Budget Speech |
1-5/2/2008 |
2-4/2/2009 |
29/1-2/2/2010 |
7-11/2/2011 |
3-6/1/2012 |
1-6/2/2013 |
4-6/2/2014 |
30/1-4/2/2015 |
1-4/2/2016 |
||||||||
|
FS’s rating before the Budget & error [13] |
56.0+/-1.2 |
56.7+/-1.1 |
58.3+/-1.1 |
55.4+/-1.7 |
50.6+/-1.6 |
57.8+/-1.4 |
56.7+/-1.4 |
58.6+/-1.8 |
62.3+/-1.5 |
||||||||
|
FS’s net approval rate before the Budget & error [13] |
24+/-4% |
32 +/-5% |
46 +/-4% |
33+/-6% |
13+/-7% |
45+/-5% |
33+/-6% |
42+/-6% |
51+/-5% |
||||||||
|
Change in
|
+11.9 [15] |
-1.8 [15] |
+3.0 [15] |
-3.0 [15] |
+3.5 [15] |
-1.2 |
-2.7 [15] |
+2.4 [15] |
-0.1 |
||||||||
|
Change in FS’s
|
+30% [15] |
-4% |
-- |
-20% [15] |
-10% [15] |
-10% [15] |
-6% |
+2% |
-3% |
||||||||
|
Popularity of Paul Chan |
|||||||||||||||||
|
Year of the Budget |
2017-18 |
2018-19 |
|||||||||||||||
|
Date of Budget Speech |
22/2/2017 |
28/2/2018 |
|||||||||||||||
|
FS’s rating at Budget instant survey & error [13] |
47.4+/-2.3 |
44.5+/-2.3 |
|||||||||||||||
|
FS’s net approval rate at Budget instant survey & error [13] |
4+/-6% |
-12+/-7% |
|||||||||||||||
|
Date of the latest survey before Budget Speech |
6-9/2/2017 |
1-6/2/2018 |
|||||||||||||||
|
FS’s rating before the Budget & error [13] |
34.0+/-2.1 |
44.3+/-1.8 |
|||||||||||||||
|
FS’s net approval rate before the Budget & error [13] |
-29+/-6% |
-12+/-6% |
|||||||||||||||
|
Change in FS’s rating |
+13.4 [15] |
+0.2 |
|||||||||||||||
|
Change in FS’s net approval rate |
+33% [15] |
-- |
|||||||||||||||
[11] FS rating was introduced in our Budget instant poll in 2001, while approval rate was introduced in 2005. This table therefore starts from 2001.
[12] These questions only use the concerned sub-samples for the tracking surveys between 2006 and 2010.
[13] All errors 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. Questions using only sub-samples would have bigger sample error. Sampling errors of ratings are calculated according to the distribution of the scores collected.
[14] The frequency of FS rating and approval rate was different before November 2005.
[15] 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 is not the same as whether they are practically useful or meaningful, and different weighting methods could have been applied in different surveys.
The survey showed that the latest popularity rating of FS Paul Chan after his Budget Speech was 44.5 marks, with an approval and disapproval rates of 30% and 41% respectively, giving a net approval of negative 12 percentage points.
Indepth Analysis
In the survey, we also asked respondents for their age. If they were reluctant to give their exact age, they could give us a range. According to their answers, we grouped them into 18-29, 30-49, and 50 years or older. Herewith further analysis of public’s rating and satisfaction rate of the Budget, and Paul Chan’s approval rates by respondents’ age:
|
Date of survey: 28/2/2018 |
18-29 |
30-49 |
50+ |
Overall |
|
|
Rating of the Budget[16] |
39.9+/-4.6
|
47.8+/-3.5
|
51.9+/-3.3
|
48.3+/-2.2
|
|
|
Satisfaction rate of the Budget[16] |
Satisfaction |
9+/-6%
|
24+/-6%
|
35+/-6%
|
26+/-4%
|
|
Half-half |
27+/-9%
|
32+/-7%
|
26+/-5%
|
28+/-4%
|
|
|
Dissatisfaction |
58+/-10%
|
39+/-7%
|
34+/-6%
|
40+/-4%
|
|
|
Don’t know / hard to say |
6+/-5%
|
6+/-3%
|
4+/-3%
|
5+/-2%
|
|
|
Total |
100%
|
100%
|
100%
|
100%
|
|
[16] Differences among sub-groups are tested to be statistically significant at 99% confidence level.
|
Date of survey: 28/2/2018 |
18-29 |
30-49 |
50+ |
Overall |
|
|
Vote of confidence/ no confidence in FS Paul Chan [17] |
Support |
12+/-6%
|
26+/-6%
|
39+/-6%
|
30+/-4%
|
|
Oppose |
59+/-10%
|
40+/-7%
|
36+/-6%
|
41+/-4%
|
|
|
Don’t know / hard to say |
30+/-9%
|
34+/-7%
|
25+/-5%
|
29+/-4%
|
|
|
Total |
100%
|
100%
|
100%
|
100%
|
|
[17] Differences among sub-groups are tested to be statistically significant at 99% confidence level.
Commentary
Frank Lee, Research Manager of POP, observed, “Our latest Budget instant survey shows that, among those who had some knowledge of the second Budget Speech by FS Paul Chan, 26% said they were satisfied, 41% said they were not, giving a net satisfaction rate of negative 14 percentage points, which goes down significantly by 29 percentage points from his first Budget last year. On a scale of 0-100, this Budget scored 48.2 marks, which is also significantly worse than that of last year, down by 7.5 marks. The rating of the instant survey shows that people’s appraisal of this Budget has reached the record low since the survey question was first asked in 2008. As for Paul Chan’s own popularity after the Budget, both his support rating and net approval rate have not changed much since early February, meaning that the Budget has not much instant impact on his popularity. Further analyses show that the younger the respondents, the less satisfied they are with the Budget and less supportive towards Paul Chan. Our instant survey describes people’s instant reaction towards the Budget, how people’s reaction will change after knowing more about the Budget will be revealed by our next follow-up survey.”
Future Release (Tentative)
