#Tabout stata full#

Using if in the subpop option does notВ remove cases from the analysis. It shouldВ be stressed that this is VERY different from using if to remove casesВ from an analysis. You can also use if when defining your subpopulation. Number of PSUs = 610 Population size = 6094.03 Note: subpop() takes on values other than 0 and 1

Total | 620 100.00 svy, subpop(both1): mean ell Notice the note that Stata provides whenВ the subpopulation variable is not coded 0/1. The output of the svy: meanВ command shows that the all of the cases not coded 0 or missing (the 424 casesВ coded as 2) areВ included in the subpopulation. The output of the tab command shows usВ that the recoding went as planned. WeВ will also set some values to missing, to see what happens with missing values inВ the subpopulation variable. Now letвЂ™s create a copy of both and recode the 1s to 0s. Subpop() = 0 indicates observation not in subpopulation Subpop() = 1 indicates observation in subpopulation

#Tabout stata full#

Note: subpop() subpopulation is same as full population Because we have no cases coded as 0, allВ of the cases are included in the subpopulation, as explained in the note in theВ output. Non-0 values are included in the analysis, except for missing values, whichВ are excluded from the analysis. As stated inВ the Stata SurveyВ manual, when the subpop option is used, the subpopulation is actuallyВ defined by the 0s (false), which indicate those cases to be excluded from theВ subpopulation. This is because the subpop optionВ must have a true/false variable. However, in the output of the svy: mean command, we see that all of the observations, 6194 cases, areВ included in the subpopulation. We willВ want to know this later on.) Notice in the output of the svy: tabВ command that there are 1888 cases coded 1. (The missing option is usedВ here to show that there are no missing values for this variable. Now letвЂ™s try to use a variable coded 1/2 instead of 0/1. (It is not a whole number because weВ are estimating this value using the probability weights.) In the output ofВ the svy: mean command, we also see that 789.552 cases are included in theВ subpopulation. (The missing option is used here to show that there are no missing values for this variable.В We will want to know this later on.) Notice in the output of the svy: tabВ command that there are 789.6 cases coded 1. Here we can see that yr_rnd is coded 0/1. Number of PSUs = 620 Population size = 6194 As you will see, the subpop option handles these twoВ variables differently. Next, we will consider two variables to use with the subpop option, yr_rnd, which is coded 0/1, and both, whichВ is coded 1/2. We will start by looking at the mean of our continuousВ variable, ell. However, the subpop and overВ options work the same for all svy commands. Rao (2003).ДЇor the sake of consistenc y, we will use the meanВ command for all of our examples. Cochran (1977) and Small Area Estimation by J. For more information onВ this issue, please see SamplingВ Techniques, Third Edition by William G. If the data set is subset, meaning thatВ observations not to be included in the subpopulation are deleted from the dataВ set, the standard errors of the estimates cannot be calculated correctly.В When the subpopulation option(s) is used, only the cases defined by theВ subpopulation are used in the calculation of the estimate, but all cases areВ used in the calculation of the standard errors. Using the subpopulation option(s) is extremely important whenВ analyzing survey data.

Stata has two subpopulation options that are very flexible andВ easy to use. WhenВ analyzing these subpopulations (AKA domains), you need to use the appropriateВ option. When analyzing survey data, it is common to want to look only a certainВ respondents, perhaps only women, or only respondents over age 50.