Research Method返回

All existing members of the Hong Kong Society of Accountants (HKSA) were invited to participate in this survey. Access to target respondents' contacts was restricted to the Office of Legislative Councillor Eric Li, which was responsible for printing and distributing the questionnaires. Operationally, printed questionnaires were sent to all target subjects under the cover of "Eric's Bits and Bytes, May 2004", and as an insert to the monthly newsletter of HKSA. A total of 22, 400 questionnaires were sent, according to the Office of Eric Li.
 
To avoid repetitive submissions from the same respondent, one set of unique computer-generated serial number and password was printed at the beginning of each questionnaire. Such information was solely used for verification purpose, not as a means to match any individual's identity.
 
Upon receiving the questionnaire, respondents were given the liberty to choose from 3 return modes: (1) to complete the paper questionnaire and return it by post using the pre-paid postage envelope, (2) to complete the paper questionnaire and return it by fax to POP's office, or (3) to fill in the survey online by logging onto the project website at http://hkupop.hku.hk/accountant using the unique serial number and password printed on the questionnaire.
 
The survey questionnaire comprised two parts, the first part consisted of 7 opinion questions, of which 6 were closed-ended questions and 1 was open-ended question, while the second part consisted of 7 personal information questions.
 
A total of 316 returns were received and the verification procedures for the submissions were as follows:
 
Figure 1. Verification procedures for submissions
Figure 1.  Verification procedures for submissions
 
Out of the 299 valid returns, 178 respondents (59.5%) returned the questionnaire by local mail, 20 (6.7%) by fax, and 101 (33.8%) returned them through online submission.
 
To sum up, since 299 valid questionnaires were used for the final analysis, the effective response rate was 1.33%, as shown in the following calculation (Table 1). The standard sampling error for percentages based on this sample was less than 2.9 percentage points. In other words, the sampling error for all percentages was less than plus/minus 5.8 percentage points at 95% confidence level. However, due to the very low response rate, significant systematic bias was bound to have occurred, and the findings reported herewith would not be representative of all members of the Hong Kong Society of Accountants.
 

Table 1 - Calculation of overall response rate

   Overall response rate
= (Final valid cases / Total number of questionnaires distributed) x 100%
= (299 / 22,400) x 100%
= 1.33%