Page 103 - TAIWAN TOBACCO CONTROL ANNUAL REPORT 2019

household who meet relevant criteria is unequal, residential phone surveys are characterized by unequal probability. In addition, interview failure rates are gradually increasing (especially among young people). The "Hung’s" Household sampling method has therefore been flexibly adjusted and a weighted household method has been adopted to increase the likelihood of selection of young people (not every household member is equally likely to be selected). This helps make up for the shortcomings of traditional interview methods and enhances sample representativeness. A flexible household sampling method is adopted to minimize inference bias. Collected data would then be checked and cleaned to remove any errors and undergo logical verification. To ensure that the data is capable of reflecting population characteristics and to provide a clear understanding of long- term trends of smoking prevalence in Taiwan and Fujian provinces, population statistics at the end of 2000 provided by the Directorate-General of Budget, Accounting, and Statistics (DGBAS) were used to conduct weighted analysis of the statistics against the population. To prevent from aging population affect the trend of smoking rate older people tend to stop smoking for certain diseases occur. The data was weighted with raking method by adjusting gender, age, education background, and county and city administrative districts, until there was no significant difference between the sample structure and the population structure. The adjusted sample data of gender, age, education background, as well as county and city distribution should not exhibit any significant differences with those of the population. The primary items for this investigation survey included smoking behavior, smoking cessation behavior, frequency of exposure to second-hand smoke, and awareness of smoking cessation services offered by healthcare and medical agencies of the general public. Hence, in addition to monitoring changes to smoking behaviors in Taiwan, the HPA also carried out cross-over analysis of demographic variables and socio-economic status of the survey respondents. Results could then be provided to the government as a reference for establishing future policies. 101 Research, Monitoring, and International Exchange