proc surveymeans std error of mean and sum = 0

[chopper]

I have the following SAS statement that yields a 0 for std error of mean and sum. I am trying to understand why. Any help would be appreciated. I was able to get output if I set varmethod to jackknife. Does that help?

Thanks.

data sales_str_sizes;
input str_sales _total_;
datalines;
1 425
2 20
3 6
4 2
5 2
6 2
7 43
;
proc surveymeans data=sales_neyman mean sum clsum clm total=sales_str_sizes;
	var sales;
	weight SamplingWeight;
	strata str_sales;
	title "Neyman Allocation for Sales";
run;

[ballardw]

The answer probably lies in the data summarized.

Note that part of the reason to use a TOTAL secondary set is to generate a finite population correction but this is not done if you ue BRR or jacknife. So that would be part of the reason you get a different result with jackknife: no correction.

Having strata with a population of 2 such as your strata 4, 5 and 6 looks a tad odd to me.

[chopper · Accepted solution]

The problem was my strata totals were for the sample and not the population.