How to obtain cumulative frequencies with the command "svy" ?

[Michael Duarte Goncalves]

Hello everyone,
Does anyone knows how can I add the proportions up to the point they sum to 1 please, which obtained through the command:

svy,subpop(c'i'):prop k_uebertritt2

Basically, I want to obtain cumulative frequencies in order to plot them afterwards in a twoway graph bar.
Thanks in advance!

Michael

[Leonardo Guizzetti]

The command to use would be -tab-, with the survey prefix there is also an option for confidence intervals. I am not at my computer and I can't quite recall if it will give cumulative frequencies/proportions or if you need to derive them with its output. Therefore, I suggest to give -help svy tab- a close read. Then again, these are usually less interesting than group-specific frequencies/proportions.

[David Radwin]

If you'd rather convert percentages to proportions than calculate cumulative frequencies, and you aren't plotting confidence intervals or other measures of variance, you can get cumulative percentages using aweights like the final command below:

. webuse nhanes2

. svyset

Sampling weights: finalwgt
             VCE: linearized
     Single unit: missing
        Strata 1: strata
 Sampling unit 1: psu
           FPC 1: <zero>

. svy: tab race
(running tabulate on estimation sample)

Number of strata = 31                            Number of obs   =      10,351
Number of PSUs   = 62                            Population size = 117,157,513
                                                 Design df       =          31

----------------------
     Race | proportion
----------+-----------
    White |      .8792
    Black |      .0955
    Other |      .0253
          | 
    Total |          1
----------------------
Key: proportion = Cell proportion

. tab race [aweight=finalwgt]

       Race |      Freq.     Percent        Cum.
------------+-----------------------------------
      White |9,100.12772       87.92       87.92
      Black | 988.581772        9.55       97.47
      Other | 262.290507        2.53      100.00
------------+-----------------------------------
      Total |     10,351      100.00

[Michael Duarte Goncalves]

Hello David,

Thanks for the suggestion. Very interesting in fact.

Michael