Dear R People I´m student of master in statistic and data analysis. I did use R (SURVEY Package) to estimate sampling variance. The data is come from Incoming and Expendure Survey in Maputo City (Mozambique) careout by National Statistic Institute of Mozambique, from June 2002 to July 2003 . That is a stratified multi stage sampling design, with four strata (H=1, 2, 3 and 4). Within is strata an independent sample of PSU was selected and, in each sample PSU, one SSU (Enumeration Area-EA) was selected with PPS. Within each SSU 12 household was selected with igual probability. In the Mozambican context, the PSU is a group of 3-5 EA´s; one EA is a geographical area with 100-150 households. So the 2002/2003 Incoming and Expendure survey in Maputo City is a stratified multi stage sampling Survey: with 4 stratas and within each strata with three stage of selection (first, selection of PSU, second, selection of EA and third, selection of 12 household in each sample EA). I know, according to the Theory that in this kind of samples, the resample methods of variance estimation are used at the PSU level, assumpting that the PSU´s were selected with replacement. My sample size is 78 PSU, 78 EA and 936 households (936=78x12). My estimators are all Horvitz Thompson. So, with this knowledge, i did use SURVEY package: 1º I did define the stratified 2 level cluster design in Survey package; 2º From that sampling plan, i did define the replicate design for Bootstrap with 500 replicates. I would like to know, for example, in the first strata with 20 PSU how many subsamples or replicates are in default in SURVEY Package. My Regards Carlos
______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
