Research DesignBack

The target population of this survey was all local SMEs, defined as local companies with annual sales turnover up to HK$100 million, and MMEs, defined as local companies with annual sales turnover over HK$100 million, covering many different sectors in Hong Kong except those in the service industry such as financing, insurance, legal, public relation, advertising, etc.

This was a telephone survey conducted by interviewers under close supervision. The contacts were randomly drawn from the Dun & Bradstreet HK Database, supplemented by the SME Directory maintained by the Hong Kong Trade Development Council. Duplicated numbers were filtered, and the remaining numbers were mixed in random order to produce the final sampling frame.

Data were collected by interviewers using a Web-based Computer Assisted Telephone Interview (Web-CATI) system which allowed real-time data capture and consolidation. When telephone contact was successfully established with a target company, the employer or one representative who was in a position to make business decision for that company was then invited to participate in this survey. Each contact was called back 5 times before the number was dropped as 「non-contact」, and appointments would be made if the target respondent was not available at the time we called. All interviews were conducted anonymously.

The fieldwork was conducted during the period of January 18 – February 26, 2010. A total of 1,005 representatives from the local MMEs and SMEs who spoke Cantonese were successfully interviewed. Among them, 303 respondents were from MMEs and 702 were from SMEs. As shown from the calculation in Appendix I, the overall effective response rate of this survey was 62.2% (Table 2), and the standard sampling error for percentages based on this sample was less than 1.6 percentage points. In other words, the sampling error for all percentages based on the full sample was less than plus/minus 3.2 percentage points at 95% confidence level, while the sampling error for the MME and SME sub-samples was less than plus/minus 5.7 and 3.8 percentage points respectively.