How to get a weighted count of observations using sum

[Wally Knox]

I want to use sum to get a weighted count of observations. I can weight sum, but the memory, r(N), is unweighted. How can I get a memory of a weighted count?

[Clyde Schechter]

I think you need to show us the exact code you tried and the exact output you got from Stata (including the results of -return list-). I can't replicate your problem:

. sysuse auto, clear
(1978 Automobile Data)

. count
  74

. gen wt = rpoisson(3)

. summ price [fweight = wt]

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
       price |        214    5839.346    2643.194       3291      15906

. return list

scalars:
                  r(N) =  214
              r(sum_w) =  214
               r(mean) =  5839.345794392523
                r(Var) =  6986474.85639068
                 r(sd) =  2643.194063323895
                r(min) =  3291
                r(max) =  15906
                r(sum) =  1249620

.

[Wally Knox]

Thanks for the nudge Clyde. Below is how I corrected what I was doing. I was using data from IPUMS and using their "perwt" as the weighting variable but I had not classified the weight as an fweight. Once I did that it produced an estimate of the population statistic. Before weighting the N was 2718. After fweighting it was 308381. FYI I am counting the number of people working in the auto industry (ind1950==376) in 1950. Thanks again.

gen numm = _n;

sum numm;

sum numm if ind1950==376;

gen tot376 = r(N);

display tot376;

sum numm if ind1950==376 [fweight=perwt];

gen weighttot376 = r(N);

display tot376;

display weighttot376;


Variable | Obs Mean Std. Dev. Min
> Max
-------------+-----------------------------------------------
> ---------
numm | 143559 71780 41442.06 1
> 143559

Variable | Obs Mean Std. Dev. Min
> Max
-------------+-----------------------------------------------
> ---------
numm | 2718 69548.55 41451.88 176
> 143356
2718

Variable | Obs Mean Std. Dev. Min
> Max
-------------+-----------------------------------------------
> ---------
numm | 308381 69279.89 41776.65 176
> 143356
2718
308381