Sample Design
This research program is unique in that the survey sample covers -- and represents with statistical precision -- the entire adult population (age 18 and older) of the People's Republic of China with the exception of Macau and Hong Kong SAR.
A total of 3,597 hour-long, in-person, in-home interviews were conducted across China -- 3,074 during June and July, and an additional 523 in November 2004. Interviewing was conducted throughout the country -- that is, across each of China’s 22 provinces, 3 municipalities, and 5 autonomous regions.
The strict probability-based sample was drawn in the following stages:
1. 2,500 counties, cities, and urban districts were divided into 50 strata based on their geographic location, degree of economic development, and proportion of non-agricultural population.
2. One primary sampling unit (PSU), consisting of either a county or a city, was selected from each stratum based on probability proportional to population size.
3. Within each PSU, all neighborhoods and villages were compiled. From this list, four neighborhoods or villages were selected proportional to size.
4. From each of these four neighborhoods or villages, five households were selected at random.
5. One respondent was selected from each of the selected households according to the Kish method. This research procedure, designed to ensure proper representation of all age groups and both genders in the sample, involves first recording the ages and sex of each of a selected household's adults on a grid. The respondent to be interviewed is then selected according to a prescribed systematic procedure.
6. If the designated respondent was not at home, or could not be reached, a second -- or, if needed, a third time -- an adult family member was selected systematically from among the household members remaining on the grid listing. If contact with the designated respondent could not be made after a total of three separate visits to the household, an interview with a respondent in a substitute household in the same locality was permitted. (Two substitute households were kept in reserve for each five assigned households in the interviewing area.)
The same basic selection procedure, based on probability of selection according to population size, was also used for each of the augmented ("oversampled") 10-city sub-samples.
In Beijing, Shanghai, and Guangzhou, a minimum of 400 interviews was conducted (Beijing: 402, Shanghai: 422, Guangzhou: 411: three city total: 1,235). In “oversampled” cities 4 through 10 (Chongqing, Tianjin, Shenyang, Wuhan, Nanjing, Chengdu, and Xian), an additional 807 interviews were conducted.
Among other areas classified as urban, a further 523 interviews were conducted, thus bringing the sum of all urban sector interviews to 2,565. Finally, 1,032 interviews were conducted in areas classified as rural.
To correct for the effects of deliberate oversampling of urban areas, post-weighting has been applied so that the final data are statistically accurate within plus or minus 2 percentage points and projectable to the total adult population of China (excluding Hong Kong and Macau).
