Sampling Design for 2010 CGSS,Yongjin Jin School of Statistics Renmin University of China,Email: Website: ,Outline,Background,1,Discussion,2,Sampling Design,3,The Final Result,4,Outline,Background,1,Discussion,2,Sampling Design,3,The Final Sample,4,Background,1,To fully understand social development of urban and rural areas,2,To make comparisions between urban and rural groups,1. Survey Objectives,Background,1,To fully understand social development of urban and rural areas,2,To make comparisons between urban and rural groups,3,To reflect regional discrepancy of the development in China,1. Survey Objectives,Background,2. Target Population,Three-Stage stratified probability sampling A respondents selected randomly from the ultimate unit (household),3. Sampling Method,All the household in 31 provinces (excluding Hong Kong, Macao and Taiwan),Outline,Background,1,Discussion,2,Sampling Design,3,The Final Sample,4,Discussion,Category,Category 1 Certainty unite Self-representative 必选层,Category 2 Selected unite 抽选层,Discussionstratification,Self-representative,Important unites 36 Municipalities, provincial capitals and sub-provincial cities.,Discussionstratification,Self-representative,36 Municipalities, provincial capitals and sub-provincial cities.,Consideration,Economic level Educational level Degree of openness,Discussionstratification,Self-representative,36 Municipalities, provincial capitals and sub-provincial cities.,index,GDP Number of teachers FDI Actual used foreign investment,Consideration,Economic level Educational level Degree of openness,Discussionstratification,Self-representative,36 Municipalities, provincial capitals and sub-provincial cities.,index,GDP Number of teachers FDI Actual used foreign investment,Consideration,Economic level Educational level Degree of openness,Top Five： Beijing, Shanghai, Guangzhou, Shenzhen, Tianjing,factor analysis,Discussionstratification,selected category,Other regions excluding the self-representatives,Discussionstratification,selected category,Other regions excluding the self-representatives,Consideration,District level（ urban） and county level（ rural） Futher stratification,Discussionstratification,selected category,Other regions excluding the self-representatives,index,Density of population Proportion of non-agricultural population GDP per capita,Consideration,District level（and country level,Discussionstratification,selected category,Other regions excluding the self-representatives,index,Density of population Proportion of non-agricultural population GDP per capita,Consideration,District level and country level,District level : 19 stratum Country level : 31 stratum,cluster analysis,Sampling stageAdministrative Divisions of China,Discussionstage,Discussionstages,Self-representative,Subdistrict or street,Neighborhood committee,household,selected stratum,District- or county-level units,Neighborhood committee Villagers committee,household,Discussionsample size and allocation,Total sample size: 12 000 units,Discussionsample size and allocation,Total sample size: 12 000 units,Operational definition,Neighborhood committee consisted of urban household Villagers committee consisted of rural household,Some considering,Larger discrepancy among urban individuals Urban-rural ratio,Discussionsample size and allocation,Total sample size: 12 000 units,Considering operability,Neighborhood committee consisted of urban household Villagers committee consisted of rural household,7 200 Urban household,4 800 Rural household,Considering actual situation,Larger discrepancy among urban individuals Urban-rural ratio,Discussionsample size and allocation,Stage 1,40 PSUs streets,1. Self-representative,Discussionsample size and allocation,Stage 2,Stage 1,402 SSUs Neighborhood committee,40 PSUs streets,1. Self-representative,Discussionsample size and allocation,Stage 3,Stage 2,Stage 1,40225 USUs Household（household member),402 SSUs Neighborhood committee,40 PSUs streets,1. Self-representative,40 x 2 x 25 = 2000 housholds,Discussionsample size and allocation,Stage 1,100 PSUs District, county,2. Selected stratum,Discussionsample size and allocation,Stage 2,Stage 1,1004 SSUs Neighborhood committee (Villagers committee),100 PSUs District, county,2. Selected stratum,Discussionsample size and allocation,Stage 3,Stage 2,Stage 1,100425 USUs Household（household member),1004 SSUs Neighborhood committee (Villagers committee),100 PSUs District, county,2. Selected stratum,100 x 4 x 25 = 10000 Households,Outline,Background,1,Discussion,2,Sampling Design,3,The Final Sample,4,Sampling Design,Beijing, Shanghai, Guangzhou, Shenzhen, Tianjin; The samples are all urban households; Primary sampling units (PSUs) are sub-districts in the five cities above; We use PPS to select 40 sub-districts as the PSUs in this stratum.,Sa