HKU POP releases popularity figures of CE and the Government, people’s expectation of Policy Address and ratings of the best telecommunication corporationsBack
Press Release on September 26, 2018
| Detailed Findings (Rating of Chief Executive Carrie Lam) | (People's Satisfaction with the HKSAR Government)
| Detailed Findings (People's Satisfaction with Current Political, Economic and livelihood conditions) |
| Detailed Findings (Policy Address Feature Page) |
| Detailed Findings (Ratings of the Best Telecommunication) |
Special Announcements
1. From July 2017, apart from sampling landline numbers to conduct opinion surveys, the Public Opinion Programme (POP) of The University of Hong Kong has also added mobile numbers to the sampling frame. After three months of testing, in October 2017, POP formalized the use of mixed samples as its standard for regular opinion surveys using a landline and mobile sample ratio of 4 to 1. Starting from April 2018, POP further increased the proportion of mobile sample, which the landline and mobile sample ratio became 2 to 1. The figures released today by POP have already incorporated landline and mobile samples.
2. In September 2017, POP started to use “effective response rate” to report surveys’ contact information. In July 2018, POP further revised the calculation of effective response rate. Thus, the response rates before and after the change cannot be directly compared.
3. To facilitate academic study and rational discussion, POP today released via the “HKU POP SITE” (http://hkupop.pori.hk) the raw data and related respondents’ demographics of the latest rating survey of CE Carrie Lam, together with those of regular rating surveys of former CEs CH Tung, Donald Tsang and CY Leung released earlier, for public examination. Please follow normal academic standards when using or citing such data.
Abstract
POP interviewed 1,002 Hong Kong people between 18 and 20 September 2018 by means of a random telephone survey conducted by real interviewers. Our latest survey conducted after Typhoon Mangkhut hit Hong Kong shows that the popularity rating of CE Carrie Lam has plunged by 6.5 marks to 50.8 marks, the lowest since she took office. Her latest approval rate is 40%, disapproval rate 45%, giving a net popularity of negative 4 percentage points, which has significantly decreased by 15 percentage points from two weeks ago and is the second lowest record since she took office. Indepth analysis shows that the younger and the more educated the respondents, the more critical they are of Carrie Lam as CE in terms of support rate. As for the SAR Government, its popularity has not changed much since a month ago. Its satisfaction rate now stands at 37%, dissatisfaction rate 45%, giving a net satisfaction rate of negative 8 percentage points. As for the society’s appraisals, among economic, livelihood and political conditions, people remain to be least satisfied with the current political condition. The corresponding net satisfactions are positive 2, negative 20 and negative 37 percentage points. The maximum sampling error of all percentage figures is +/-4 percentage points at 95% confidence level, while the sampling error of rating figures and net values need another calculation. The response rate of the survey is 56%.
As for the people’s expectation of Policy Address, POP conducted a double stage survey in September 2018 by means of random telephone surveys conducted by real interviewers. We used a two-stage design to study both the absolute and relative importance of different policy items. In our latest surveys, when asked to name only one issue unaided, most people considered “housing” the most pressing policy area to be handled in CE Carrie Lam’s Policy Address. In terms of absolute percentage of importance, a record high of 95% said CE should tackle housing problems, with a mean score of 4.8, meaning close to “very important”. Both figures are the highest across all items. The second to fifth items in terms of mean score are “medical policy”, “education”, “economic development” and “political development”. Four out of five items are the same as those in the last survey. Only “economic development” enters the list to replace “social welfare”. The maximum sampling error of the survey is +/-4 percentage points at 95% confidence level, response rates of the first stage and second stage survey being 50% and 56% respectively.
As for the best telecommunication corporation, POP conducted a two-stage survey in September and the results show that the most well-known telecommunication corporation was PCCW. Results of rating survey also show that PCCW has the best CSR reputation in the sector, scoring 54.5 marks, followed by Smartone and China Mobile, with 54.2 and 47.6 marks respectively. The sampling errors of rating figures are no greater than +/-2.3 marks at 95% confidence level. The response rates of the two rating surveys are 51% and 56% respectively.
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] 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.
[3] 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.
Popularity of CE and the Government
[4] The sample size of this survey is 1,002 successful interviews, not 1,002 x 55.6% response rate. In the past, many media made this mistake.
[5] “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 ratings not more than +/-1.7 and sampling error of percentages not more than +/-4%, sampling error of net values not more than +/-7% at 95% confidence level”.
People’s Expectation of Policy Address
[6] The sample size of the first stage survey is 1,030 successful interviews, not 1,030 x 50.4% response rate, while the sample size of the second stage survey is another 1,002, not 1,002 x 55.6% response rate. In the past, many media made this mistake.
[7] “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 all percentages not more than +/-4% at 95% confidence level”.
Ratings of the Best Telecommunication Corporations
[8] The sample size of the naming survey is 504 successful interviews, not 504 x 51.0% response rate, that of the first rating survey is 511 successful interviews, not 511 x 51.3% response rate, while that of the second rating survey is 1,002 successful interviews, not 1,002 x 55.6% response rate. In the past, many media made this mistake.
[9] “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 errors of ratings not more than +/-2.3 at 95% confidence level”.
I. Popularity of CE and the Government
Latest Figures
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 provided by the Census and Statistics Department. The gender-age distribution of the Hong Kong population came from “Mid-year population for 2017”, while the educational attainment (highest level attended) distribution and economic activity status distribution came from “Women and Men in Hong Kong - Key Statistics (2018 Edition)”. In the past, the mobile sample would be rim-weighted according to the basic Public Sentiment Index (PSI) figures collected in the landline sample. In July 2018, POP further refined the weighting method. The landline sample and the mobile sample would no longer be processed separately. The mobile sample would also no longer be adjusted using the basic PSI figures collected in the landline sample. The overall effect is that the importance of the mobile sample would be increased. Herewith the contact information for the latest survey:
|
Date of survey |
Sample size |
Effective response rate |
Maximum sampling error
|
|
18-20/9/2018 |
1,002 |
55.6% |
+/-3% |
[10] 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.
Recent popularity figures of CE Carrie Lam and the HKSAR Government are summarized as follows:
|
Date of survey |
3-5/7/18 |
18-23/7/18 |
6-9/8/18 |
20-22/8/18 |
3-6/9/18 |
18-20/9/18 |
Latest change |
|
Sample base |
1,001 |
1,002 |
1,000 |
1,022 |
1,030 |
1,002 |
-- |
|
Effective response rate* |
49.5% |
49.0% |
51.2% |
53.0% |
50.4% |
55.6% |
-- |
|
Finding |
Finding |
Finding |
Finding |
Finding |
Finding |
Finding & error[11] |
-- |
|
Rating of CE Carrie Lam |
55.6 |
56.2 |
55.4 |
55.3 |
57.3 |
50.8+/-1.7 |
-6.5[13] |
|
Vote of confidence in CE Carrie Lam |
49%[13] |
47% |
45% |
49% |
48% |
40+/-3% |
-8%[13] |
|
Vote of no confidence in CE Carrie Lam |
38% |
38% |
44%[13] |
35%[13] |
38% |
45+/-3% |
+7%[13] |
|
Net approval rate |
12%[13] |
8% |
1% |
14%[13] |
10% |
-4+/-6% |
-15%[13] |
|
Satisfaction rate of SARG performance[12] |
39% |
32%[13] |
-- |
37% |
-- |
37+/-4% |
-- |
|
Dissatisfaction rate of SARG performance[12] |
39% |
43% |
-- |
40% |
-- |
45+/-4% |
+4% |
|
Net satisfaction rate |
0% |
-11%[13] [14] |
-- |
-3% |
-- |
-8+/-7% |
-4% |
|
Mean value[12] |
2.9 (Base=567) |
2.7[13] [14] (Base=582) |
-- |
2.8 (Base=541) |
-- |
2.8+/-0.1 (Base=597) |
-- |
|
Trust in HKSAR Government[12] |
45%[13] |
46% |
-- |
-- |
51% |
-- |
-- |
|
Distrust in HKSAR Government[12] |
36% |
37% |
-- |
-- |
36% |
-- |
-- |
|
Net trust |
9% |
9% |
-- |
-- |
15% |
-- |
-- |
|
Mean value[12] |
3.0 (Base=562) |
3.0 (Base=583) |
-- |
-- |
3.1 (Base=507) |
-- |
-- |
|
Current economic condition:
|
43% |
-- |
-- |
40% |
-- |
40+/-3% |
-- |
|
Current economic condition:
|
33% |
-- |
-- |
33% |
-- |
37+/-3% |
+5%[13] |
|
Net satisfaction rate |
10% |
-- |
-- |
7% |
-- |
2+/-6% |
-5% |
|
Mean value[12] |
3.1 (Base=987) |
-- |
-- |
3.0 (Base=998) |
-- |
2.9+/-0.1 (Base=975) |
-0.1 |
|
Current livelihood condition:
|
31% |
-- |
-- |
29% |
-- |
30+/-3% |
+1% |
|
Current livelihood condition:
|
48% |
-- |
-- |
51% |
-- |
50+/-3% |
-2% |
|
Net satisfaction rate |
-16% |
-- |
-- |
-23% |
-- |
-20+/-5% |
+3% |
|
Mean value[12] |
2.7 (Base=992) |
-- |
-- |
2.6 (Base=1,006) |
-- |
2.6+/-0.1 (Base=986) |
-- |
|
Current political condition:
|
19% |
-- |
-- |
16% |
-- |
21+/-3% |
+4%[13] |
|
Current political condition:
|
61% |
-- |
-- |
63% |
-- |
57+/-3% |
-6%[13] |
|
Net satisfaction rate |
-42% |
-- |
-- |
-47% |
-- |
-37+/-5% |
+10%[13] |
|
Mean value[12] |
2.3 (Base=948) |
-- |
-- |
2.2 (Base=965) |
-- |
2.3+/-0.1 (Base=914) |
+0.1[13] |
[11] 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. Media can state “sampling error of rating not more than +/-1.7, sampling error of percentages not more than +/-4%, sampling error of net values not more than +/-7% at 95% confidence level” when quoting the above figures. The error margin of previous survey can be found at the POP Site.
[12] 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. This question only uses sub-samples of the tracking surveys concerned. The sample size for this series is 606.
[13] 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.
[14] Such changes have gone beyond the sampling errors at the 95% confidence level because of a change in the weighting method. If the previous weighting method was used, the changes would not have gone beyond the sampling errors.
The latest survey showed that CE Carrie Lam scored 50.8 marks, and 40% supported her as CE, her net approval rate is negative 4 percentage points. Regarding people’s appraisal of the overall performance of the HKSAR Government, the latest figures revealed that 37% were satisfied, whereas 45% were dissatisfied, thus net satisfaction stands at negative 8 percentage points. The mean score is 2.8, meaning between “half-half” and “quite dissatisfied” in general. As for people’s satisfaction with the current economic, livelihood and political conditions, the latest satisfaction rates were 40%, 30% and 21% respectively, while their net satisfaction rates in these conditions were positive 2, negative 20 and negative 37 percentage points. The mean scores of the economic, livelihood and political conditions were 2.9, 2.6 and 2.3, meaning between “half-half” and “quite dissatisfied” in general.
Indepth Analysis
In the survey, we also asked respondents for their age and education attainment. If they were reluctant to give their exact age, they could give us a range. Herewith further analysis of the support rating and support rate of Carrie Lam as Chief Executive by respondents’ age and education attainment, with sub-sample size placed in brackets:
|
Date of survey: 18-20/9/2018 |
18-29 |
30-49 |
50 or above |
Overall sample |
|
|
Rating of CE Carrie Lam[15] |
35.7+/-3.7
|
48.7+/-3.0
|
58.1+/-2.4
|
51.0+/-1.8
|
|
|
Support / Oppose Carrie Lam as CE[15] |
Support |
13+/-5%
|
36+/-5%
|
53+/-5%
|
40+/-3%
|
|
Oppose |
72+/-7%
|
49+/-5%
|
31+/-4%
|
44+/-3%
|
|
|
Don’t know / Hard to say |
14+/-5%
|
14+/-4%
|
16+/-3%
|
15+/-2%
|
|
|
Total |
100%
|
100%
|
100%
|
100%
|
|
[15] Differences among sub-groups are tested to be statistically significant at 99% confidence level.
|
Date of survey: 18-20/9/2018 |
Primary or below |
Secondary |
Tertiary or above |
Overall sample |
|
|
Support / Oppose Carrie Lam as CE[16] |
Support |
58+/-7%
|
41+/-5%
|
29+/-5%
|
40+/-3%
|
|
Oppose |
25+/-6%
|
45+/-5%
|
56+/-5%
|
45+/-3%
|
|
|
Don’t know / Hard to say |
17+/-5%
|
14+/-3%
|
16+/-4%
|
15+/-2%
|
|
|
Total |
100%
|
100%
|
100%
|
100%
|
|
[16] Differences among sub-groups are tested to be statistically significant at 99% confidence level.
Opinion Daily
In January 2007, POP opened a feature page called “Opinion Daily” at the “POP Site”, to record significant events and selected polling figures on a day-to-day basis, in order to let readers judge by themselves the reasons for the ups and downs of different opinion figures. In July 2007, POP collaborated with Wisers Information Limited whereby Wisers supplies to POP each day starting from July 24, a record of significant events of that day, according to the research method designed by POP. These daily entries would be uploaded to “Opinion Daily” as soon as they are verified by POP.
For some of the polling items covered in this press release, the previous survey was conducted from 20 to 22 August, 2018 while this survey was conducted from 18 to 20 September, 2018. During this period, herewith the significant events selected from counting newspaper headlines and commentaries on a daily basis and covered by at least 25% of the local newspaper articles. Readers can make their own judgment if these significant events have any impacts to different polling figures.
|
20/9/18 |
Media continues to report on Typhoon Mangkhut. |
|
18/9/18 |
China-US trade war intensifies. |
|
18/9/18 |
Media continues to report on Typhoon Mangkhut. |
|
17/9/18 |
The traffic is paralyzed in the first working day after Super Typhoon Mangkhut hit Hong Kong. |
|
16/9/18 |
Super Typhoon Mangkhut hits Hong Kong. |
|
7/9/18 |
Thirteen protesters who oppose North East New Territories development plan win their appeal at the Court of Final Appeal. |
|
1/9/18 |
MTR holds 2-day open days for the Express Rail Link West Kowloon Station. |
|
30/8/18 |
Office of the Ombudsman criticizes the Food and Environmental Hygiene Department for poor management of public markets. |
|
23/8/18 |
The Hong Kong Section of Express Rail Link will be commissioned on September 23. |
Online Opinion Analysis
In July 2017, POP started collaborating with uMax Data Technology Limited to conduct “Online Opinion Analysis”. uMax Data would provide technical support concerning social big data to track posts in Facebook pages and various major forums in Hong Kong that mentioned specific political figures or organizations. When public opinion changes very significantly, POP will summarize the popular online posts about the specific political figures or organizations between the two surveys and compile a list of related events. Readers could make their own judgment if the events listed have impact on the related public opinion figures.
Since the latest survey findings reveal that the net approval rate of CE Carrie Lam has changed significantly compared to the figure recorded in the last survey, POP conducted “Online Opinion Analysis” to identify the top ten most discussed issues on the internet. The list of most discussed items after eliminating duplications is as follows:
|
Carrie Lam met the press after Typhoon Mangkhut hit Hong Kong. |
|
Netizens discussed how people got to work after Typhoon Mangkhut hit Hong Kong. |
|
A pedestrian was hit by a falling tree and lost consciousness. |
|
Sammy Leung recorded a voiceover for the pseudo-trailer “Working after hurricane”. |
|
Netizens left comments on Carrie Lam’s Facebook page hoping for the government to declare a day off. |
|
Carrie Lam said it would have been irresponsible to declare a day off. |
|
Some employers deducted wages from employees who were late or absent after Typhoon Mangkhut hit Hong Kong. |
The results show that online public opinion had been focusing on CE’s decisions after Typhoon Mangkhut hit Hong Kong. Whether or not these items could sufficiently explain the changes in CE popularity ratings, readers could form their own judgment.
II. People’s expectation of Policy Address
Latest Figures
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 provided by the Census and Statistics Department. The gender-age distribution of the Hong Kong population came from “Mid-year population for 2017”, while the educational attainment (highest level attended) distribution and economic activity status distribution came from “Women and Men in Hong Kong - Key Statistics (2018 Edition)”. In the past, the mobile sample would be rim-weighted according to the basic Public Sentiment Index (PSI) figures collected in the landline sample. In July 2018, POP further refined the weighting method. The landline sample and the mobile sample would no longer be processed separately. The mobile sample would also no longer be adjusted using the basic PSI figures collected in the landline sample. The overall effect is that the importance of the mobile sample would be increased. Herewith the contact information for the latest survey:
|
Date of survey |
Sample size |
Effective response rate |
Maximum sampling error
|
|
3-6/9/2018 (First stage) |
1,030 |
50.4% |
+/-3% |
|
18-20/9/2018 (Second stage) |
1,002 |
55.6% |
+/-3% |
[17] 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.
According to our first stage survey conducted in early September 2018, when asked to name unaided one issue that CE Carrie Lam should focus on in her second Policy Address to be announced in early October, 62% of the respondents wished she would take “housing” as her first priority, while 6%, 4%, 4% and 4% respectively chose “political development”, “education”, “economic development” and “medical policy”, while 7% of the respondents failed to give a specific answer. Please refer to the “HKU POP SITE” for detailed figures.
In order to further study people’s expectation, another survey was then conducted whereby respondents were asked to evaluate each of the 5 top priority items individually, on a 5-point scale, how important it is for each item to be tackled in the Policy Address. Results compared to those of last 2 series are summarized below in descending order of mean values:
|
Date of survey |
9-12/1/17 |
3-4/10/17 |
18-20/9/18 |
Latest change |
|
Sample base |
548-632 |
523-575 |
560-624 |
-- |
|
Response rate* |
69.3% |
59.5% |
55.6% |
-- |
|
Finding |
Finding |
Finding |
Finding & error [18] |
-- |
|
Perceived housing issues as “very important” |
70% |
77%[20] |
82+/-3% |
+5%[20] |
|
Perceived housing issues as “quite important” |
15%[20] |
17% |
13+/-3% |
-4%[20] |
|
“Very” + “quite” important[19] |
85%[20] |
94%[20] |
95+/-2% |
+1% |
|
Mean value[21] |
4.5 (Base=607) |
4.7[20] (Base=518) |
4.8+/-0.1 (Base=589) |
+0.1 |
|
Perceived medical policy issues as “very important” |
62%[20] |
64% |
68+/-4% |
+5% |
|
Perceived medical policy issues as “quite important” |
22%[20] |
27%[20] |
22+/-3% |
-5%[20] |
|
“Very” + “quite” important[19] |
84% |
91%[20] |
90+/-2% |
-1% |
|
Mean value[21] |
4.4 (Base=536) |
4.5 (Base=550) |
4.6+/-0.1 (Base=606) |
+0.1 |
|
Perceived education issues as “very important” |
-- |
54% |
55+/-4% |
+1% |
|
Perceived education issues as “quite important” |
-- |
32% |
29+/-4% |
-4% |
|
“Very” + “quite” important[19] |
-- |
86% |
83+/-3% |
-3% |
|
Mean value[2+] |
-- |
4.4 (Base=566) |
4.4+/-0.1 (Base=539) |
-- |
|
Perceived economic development issues as “very important” |
50% |
-- |
45+/-4% |
-- |
|
Perceived economic development issues as “quite important” |
24% |
-- |
38+/-4% |
-- |
|
“Very” + “quite” important[19] |
74%[20] |
-- |
83+/-3% |
-- |
|
Mean value[2+] |
4.2 (Base=571) |
-- |
4.3+/-0.1 (Base=532) |
-- |
|
Perceived political development issues as “very important” |
40%[20] |
38% |
35+/-4% |
-3% |
|
Perceived political development issues as “quite important” |
23% |
31%[20] |
31+/-4% |
+1% |
|
“Very” + “quite” important[19] |
64% |
69%[20] |
66+/-4% |
-2% |
|
Mean value[21] |
3.9[20] (Base=498) |
4.0 (Base=494) |
4.0+/-0.1 (Base=487) |
-- |
|
Perceived social welfare issues as “very important” |
58%[20] |
54% |
-- |
-- |
|
Perceived social welfare issues as “quite important” |
24%[20] |
32%[20] |
-- |
-- |
|
“Very” + “quite” important[19] |
82% |
87%[20] |
-- |
-- |
|
Mean value[21] |
4.4 (Base=618) |
4.4 (Base=507) |
-- |
-- |
* “Overall response rate” was used before September 2017, thereafter,
“effective response rate” was used.
[18] 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. Media can state “sampling error of percentages not more than
+/-4% at 95% confidence level” when quoting the above figures.
[19] Percentages in these rows may not be equal to the sum of percentages shown in the rows of “very” and “quite important” due to rounding off.
[20] Such changes have gone beyond the sampling errors at the 95% confidence level under the same weighting method, 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.
[21] The mean value is calculated by quantifying all individual responses into 1, 2, 3, 4, 5 marks according to their degree of importance, where 1 is the lowest and 5 the highest, and then calculate the sample mean.
When asked to evaluate the importance of each item individually, 95% said CE Carrie Lam needed to tackle housing issue in the coming Policy Address, including 82% regarded it as “very important” and 13% “quite important”. Meanwhile, medical policy, education, economic development and political development were perceived to be important by 90%, 83%, 83% and 66% respectively. The mean scores of the five issues are 4.8, 4.6, 4.4, 4.3 and 4.0 respectively, which is between “quite important” and “very important” in general.
III. Ratings of the Best Telecommunication Corporations
In 2008, HKUPOP initiated a tracking survey series on Corporate Social Responsibility, aiming to gauge the public image of different commercial organizations in order to encourage them to become ethical companies and select the best corporations. There are a total of six modules under this survey series, namely, 1) Public Transportation, 2) Telecommunication, 3) Banks and Financial Services, 4) Real Estate and Property Development, 5) Retail, and 6) Fast Food Restaurant. From January to December 2015, the survey was sponsored by Metro Broadcast Corporation Ltd and branded as “Metro CSR Index”. Results were released every month in the website of Metro Radio.
At the beginning, these surveys were conducted once every three months, with two different modules each time. From July 2017, the frequency was changed to once every six months, with one module only for each survey. From July 2018, the frequency was further changed to once every year, with one module only for each survey. The surveys were conducted in two stages. In the first stage, respondents were requested to nominate, unprompted, at most 5 corporations that they can think of. POP would then select from this list of unprompted nominees the 3 most frequently cited names for the next stage survey. During the second stage survey, respondents would be asked to rate the CSR performance for each of the shortlisted corporations using a 0-100 scale. 0 indicates extremely poor performance, 100 indicates extremely good performance, and 50 means half-half.
Latest Figures
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 provided by the Census and Statistics Department. The gender-age distribution of the Hong Kong population came from “Mid-year population for 2017”, while the educational attainment (highest level attended) distribution and economic activity status distribution came from “Women and Men in Hong Kong - Key Statistics (2018 Edition)”. In the past, the mobile sample would be rim-weighted according to the basic Public Sentiment Index (PSI) figures collected in the landline sample. In July 2018, POP further refined the weighting method. The landline sample and the mobile sample would no longer be processed separately. The mobile sample would also no longer be adjusted using the basic PSI figures collected in the landline sample. The overall effect is that the importance of the mobile sample would be increased. Herewith the contact information for the latest survey:
|
Date of survey |
Sample size |
Effective response rate |
Maximum sampling error[22] |
|
3-4/9/2018 (Naming survey) |
504 |
51.0% |
+/-4% |
|
5-6/9/2018 (Rating survey)[23] |
511 |
51.3% |
+/-2.0 |
|
18-20/9/2018 (Rating survey)[24] |
1,002 |
55.6% |
+/-2.3 |
[22] 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 and sampling errors of ratings are calculated according to the distribution of the scores collected.
[23] This survey involved PCCW and Smartone.
[24] This survey involved China Mobile only.
The research design of our “Best Corporation” rating survey has been explained in detail under “Survey Method” in our corresponding web page. The corporations being rated in our latest survey were those which obtained highest unprompted mentions in our first stage naming survey conducted in early September 2018. In the survey, respondents could name up to 5 local telecommunication corporations which they knew best. The top three corporations mentioned most frequently in the sector were: PCCW, Smartone and China Mobile. These corporations then entered into the second stage rating survey conducted in the same month, respondents were asked to rate each short-listed corporations using a 0-100 scale. 0 indicates extremely poor performance, 100 indicates extremely good performance, and 50 means half-half.
Recent ratings of the best telecommunication corporation are summarized as follows:
|
Date of survey |
24-25/4/17 |
16-17/8/17 |
5-6/2/18 |
5-6/9/2018 / 18-20/9/2018 |
Latest change |
|
|
Sample base |
506 |
611[25] |
505 |
511 / 1,002 |
-- |
|
|
Response rate* |
74.0% |
57.6% |
61.2% |
51.3% / 55.6% |
-- |
|
|
Finding |
Finding |
Finding |
Finding |
Finding & error[26] |
Recognition rate |
-- |
|
PCCW |
51.6{1}[27] |
54.1{2}[27] |
51.7{1}[27] |
54.5+/-2.0{1} |
84.6% |
+2.8[27] [28] |
|
Smartone |
50.9{2}[27] |
-- |
51.1{2} |
54.2+/-1.9{2} |
75.3% |
+3.1[27] |
|
China Mobile |
-- |
-- |
-- |
47.6+/-2.3{3} |
56.4% |
-- |
|
Hutchison |
49.5{3}[27] |
50.7{3}[27] |
45.6{3}[27] |
-- |
-- |
-- |
|
HKBN |
-- |
58.6{1} |
-- |
-- |
-- |
-- |
* “Overall response rate” was used before September 2017, thereafter, “effective response rate” was used. In July 2018, POP revised the calculation of effective response rate. Thus, the response rates before and after the change cannot be directly compared.
[25] The mobile sample was not included when survey results were released. The figures in the table above have been updated to reflect the results based on the combined landline and mobile sample. However, whether changes have gone beyond sampling errors is still determined based on the figures in the first release.
[26] 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. Media can state “sampling error of various ratings not more than +/-2.3 at 95% confidence level” when quoting the above figures. Numbers in square brackets { } indicates rankings. The error margin of previous survey can be found at the POP Site.
[27] 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.
[28] Such changes have gone beyond the sampling errors at the 95% confidence level because of a change in the weighting method. If the previous weighting method was used, the changes would not have gone beyond the sampling errors.
Our latest survey showed that PCCW was considered as having the best CSR reputation among local telecommunication corporation, scored 54.5 marks, while Smartone and China Mobile scored 54.2 and 47.6 marks respectively.
Commentary
Note: The following commentary was written by Senior Data Analyst of POP, Edward Tai.
Our latest survey conducted after Typhoon Mangkhut hit Hong Kong shows that the popularity rating of CE Carrie Lam has plunged by 6.5 marks to 50.8 marks, the lowest since she took office. Her latest approval rate is 40%, disapproval rate 45%, giving a net popularity of negative 4 percentage points, which has significantly decreased by 15 percentage points from two weeks ago and is the second lowest record since she took office. Indepth analysis shows that the younger and the more educated the respondents, the more critical they are of Carrie Lam as CE in terms of support rate. As for the SAR Government, its popularity has not changed much since a month ago. Its satisfaction rate now stands at 37%, dissatisfaction rate 45%, giving a net satisfaction rate of negative 8 percentage points. As for the society’s appraisals, among economic, livelihood and political conditions, people remain to be least satisfied with the current political condition. The corresponding net satisfactions are positive 2, negative 20 and negative 37 percentage points. As for the reasons affecting the ups and downs of these figures, we leave it to our readers to form their own judgment using detailed records displayed in our “Opinion Daily” and “Online Opinion Analysis”.
As for people’s expectation of Policy Address, our latest survey shows that when asked to name only one issue unaided, most people considered “housing” the most pressing policy area to be handled in CE Carrie Lam’s Policy Address. In terms of absolute percentage of importance, a record high of 95% said CE should tackle housing problems, with a mean score of 4.8, meaning close to “very important”. Both figures are the highest across all items. The second to fifth items in terms of mean score are “medical policy”, “education”, “economic development” and “political development”. Four out of five items are the same as those in the last survey. Only “economic development” enters the list to replace “social welfare”.
As for the best telecommunication corporation, our survey conducted in September shows that the most well-known telecommunication corporation was PCCW. Results of rating survey also show that PCCW has the best CSR reputation in the sector, scoring 54.5 marks, followed by Smartone and China Mobile, with 54.2 and 47.6 marks respectively.
Future Release (Tentative)
Public Sentiment Index
Reference – Technical Notes of “Online Opinion Analysis”
Technically, POP research team firstly determines and inputs related keywords to be searched on the platform provided by uMax Data. POP then selects targeted online platforms to be monitored (currently selected: Facebook and forums), and extract the top five most discussed items on each online platform, thus come up with a list of 10 items in total. By eliminating duplications in the top ten items, POP compiles a table which is included in the press release for readers’ reference.
