Research Design返回


Research design


The target population of this part was Cantonese-speaking population of Hong Kong aged 15 or above. To minimize sampling bias, the following sampling technique was adopted:


Telephone numbers were first drawn randomly from the residential telephone directories as "seed numbers", from which another set of numbers was generated using the "plus/minus one/two" method, in order to capture the unlisted numbers. Duplicated numbers were then filtered, and the remaining numbers were mixed in random order to produce the final telephone sample.


When telephone contact was successfully established with a target household, one person aged 15 or above was selected. If more than one subject had been available, selection was made using the "next birthday rule" which selected the person who had his/her birthday next from all those present.


Telephone interviews were carried out from 1830 to 2230 hours between 28 September and 5 October 2001. Data were collected by interviewers using a Computer Assisted Telephone Interviews (CATI) system under close supervision.


As shown from the detailed breakdown of the contact information, amongst the 7,678 telephone numbers sampled for the survey, 889 were found to be invalid lines, 2,309 were non-contacts, including busy lines and no answer, while another 306 were business lines, 584 were fax lines or with answering machines. Of the 3,590 valid contacts, the numbers of refusals (without knowing whether target respondents were qualified or not), disqualified and unfinished cases with appointment dates beyond the end of fieldwork period were 808, 91 and 275 respectively. Besides, 61 were incomplete due to unexpected termination of interviews, 800 have language and other problems, and the remaining 1,555 were successful cases.

Table 1. Breakdown of contact information of the survey
   Frequency  Percentage
 Telephone numbers attempted  7,678   
 Invalid numbers  889  
 Busy / no answer  2,309  
 Business lines  306  
 Fax / answering machine  584  
 Valid contacts  3,590   100.0% 
 Refusal  808  22.5%
 Failed to qualify  91  2.5%
 Appointment  275  7.7%
 Incomplete  61  1.7%
 Other problems  800  22.3%
 Successful  1,555  43.3%

To sum up, a total of 1,555 Hong Kong residents aged 15 or above were successfully interviewed in this survey. The effective response rate was 64.2% as shown in the following calculation. The standard sampling error for percentages based on this sample was less than 1.3 percentage points.

Table 2 Calculation of effective response rate

   Effective response rate
= [Successful cases / (Successful cases + Refusal cases +Incomplete cases)] 100%
= [1,555 / (1,555 + 808+ 61)] 100%
= 64.2%

To improve the representativeness of the results, all data in the first part of this study have been weighted according to the age and gender distributions of the Hong Kong population as reported in the 2001 Population Census.