st: Setup for Survey Sampling--Example 9.4 from Scheaffer

[Susan Cochran <cochran@nicco.sscnet.ucla.edu>]

Hi, all,

Can you help?  I can't get this to produce the right estimate.

I am trying to get Stata 9 to reproduce the analysis of Table 9.1 In 
Scheaffer et al., 6th edition, p. 307.

There are 90 plants, 10 are sampled SRS without replacement at the first 
stage, and within plants machines (m) are sampled SRS without replacement 
and measured for the hours of being broken.  The known population size is 
about 4500 machines.  M=number of machines in the plant sampled.

The calculations by hand reveal a mean of 4.8 hours.

I created a raw data file with the following structure

plant 	M 	m 	hours 	nplant 	nmach	p1 	p2 	pwt
1	50	10	5	90	4500	9.00	5	45
1	50	10	7	90	4500	9.00	5	45
1	50	10	9	90	4500	9.00	5	45
1	50	10	0	90	4500	9.00	5	45
1	50	10	11	90	4500	9.00	5	45
1	50	10	2	90	4500	9.00	5	45
1	50	10	8	90	4500	9.00	5	45
1	50	10	4	90	4500	9.00	5	45
1	50	10	3	90	4500	9.00	5	45
1	50	10	5	90	4500	9.00	5	45
2	65	13	4	90	4500	9.00	5	45
2	65	13	3	90	4500	9.00	5	45
2	65	13	7	90	4500	9.00	5	45
2	65	13	2	90	4500	9.00	5	45
2	65	13	11	90	4500	9.00	5	45
2	65	13	0	90	4500	9.00	5	45
2	65	13	1	90	4500	9.00	5	45
2	65	13	9	90	4500	9.00	5	45
2	65	13	4	90	4500	9.00	5	45
2	65	13	3	90	4500	9.00	5	45
2	65	13	2	90	4500	9.00	5	45
2	65	13	1	90	4500	9.00	5	45
2	65	13	5	90	4500	9.00	5	45
3	45	9	5	90	4500	9.00	5	45
3	45	9	6	90	4500	9.00	5	45
3	45	9	4	90	4500	9.00	5	45
3	45	9	11	90	4500	9.00	5	45
3	45	9	12	90	4500	9.00	5	45
3	45	9	0	90	4500	9.00	5	45
3	45	9	1	90	4500	9.00	5	45
3	45	9	8	90	4500	9.00	5	45
3	45	9	4	90	4500	9.00	5	45
4	48	10	6	90	4500	9.00	4.8	43.2
4	48	10	4	90	4500	9.00	4.8	43.2
4	48	10	0	90	4500	9.00	4.8	43.2
4	48	10	1	90	4500	9.00	4.8	43.2
4	48	10	0	90	4500	9.00	4.8	43.2
4	48	10	9	90	4500	9.00	4.8	43.2
4	48	10	8	90	4500	9.00	4.8	43.2
4	48	10	4	90	4500	9.00	4.8	43.2
4	48	10	6	90	4500	9.00	4.8	43.2
4	48	10	10	90	4500	9.00	4.8	43.2
5	52	10	11	90	4500	9.00	5.2	46.8
5	52	10	4	90	4500	9.00	5.2	46.8
5	52	10	3	90	4500	9.00	5.2	46.8
5	52	10	1	90	4500	9.00	5.2	46.8
5	52	10	0	90	4500	9.00	5.2	46.8
5	52	10	2	90	4500	9.00	5.2	46.8
5	52	10	8	90	4500	9.00	5.2	46.8
5	52	10	6	90	4500	9.00	5.2	46.8
5	52	10	5	90	4500	9.00	5.2	46.8
5	52	10	3	90	4500	9.00	5.2	46.8
6	58	12	12	90	4500	9.00	4.83	43.5
6	58	12	11	90	4500	9.00	4.83	43.5
6	58	12	3	90	4500	9.00	4.83	43.5
6	58	12	4	90	4500	9.00	4.83	43.5
6	58	12	2	90	4500	9.00	4.83	43.5
6	58	12	0	90	4500	9.00	4.83	43.5
6	58	12	0	90	4500	9.00	4.83	43.5
6	58	12	1	90	4500	9.00	4.83	43.5
6	58	12	4	90	4500	9.00	4.83	43.5
6	58	12	3	90	4500	9.00	4.83	43.5
6	58	12	2	90	4500	9.00	4.83	43.5
6	58	12	4	90	4500	9.00	4.83	43.5
7	42	8	3	90	4500	9.00	5.25	47.25
7	42	8	7	90	4500	9.00	5.25	47.25
7	42	8	6	90	4500	9.00	5.25	47.25
7	42	8	7	90	4500	9.00	5.25	47.25
7	42	8	8	90	4500	9.00	5.25	47.25
7	42	8	4	90	4500	9.00	5.25	47.25
7	42	8	3	90	4500	9.00	5.25	47.25
7	42	8	2	90	4500	9.00	5.25	47.25
8	66	13	3	90	4500	9.00	5.0769	45.69
8	66	13	6	90	4500	9.00	5.0769	45.69
8	66	13	4	90	4500	9.00	5.0769	45.69
8	66	13	3	90	4500	9.00	5.0769	45.69
8	66	13	2	90	4500	9.00	5.0769	45.69
8	66	13	2	90	4500	9.00	5.0769	45.69
8	66	13	8	90	4500	9.00	5.0769	45.69
8	66	13	4	90	4500	9.00	5.0769	45.69
8	66	13	0	90	4500	9.00	5.0769	45.69
8	66	13	4	90	4500	9.00	5.0769	45.69
8	66	13	5	90	4500	9.00	5.0769	45.69
8	66	13	6	90	4500	9.00	5.0769	45.69
8	66	13	3	90	4500	9.00	5.0769	45.69
9	40	8	6	90	4500	9.00	5	45
9	40	8	4	90	4500	9.00	5	45
9	40	8	7	90	4500	9.00	5	45
9	40	8	3	90	4500	9.00	5	45
9	40	8	9	90	4500	9.00	5	45
9	40	8	1	90	4500	9.00	5	45
9	40	8	4	90	4500	9.00	5	45
9	40	8	5	90	4500	9.00	5	45
10	56	11	6	90	4500	9.00	5.09    45.818
10	56	11	7	90	4500	9.00	5.09	45.818
10	56	11	5	90	4500	9.00	5.09	45.818
10	56	11	10	90	4500	9.00	5.09	45.818
10	56	11	11	90	4500	9.00	5.09	45.818
10	56	11	2	90	4500	9.00	5.09	45.818
10	56	11	1	90	4500	9.00	5.09	45.818
10	56	11	4	90	4500	9.00	5.09	45.818
10	56	11	0	90	4500	9.00	5.09	45.818
10	56	11	5	90	4500	9.00	5.09	45.818
10	56	11	4	90	4500	9.00	5.09    45.818



When I specified the following set up

svyset plant [pweight=pwt], fpc(nplant) vce(linearized) || _n, fpc(M)

The total calculated correctly, as did the SE, but the mean is incorrect 
(showing the simple mean of the dataset  4.6 not the mean of 4.8 which is 
correct).  This is because (?) the population size is seen as 4698  (the 
sum of the weights) not 4500 and the total hours/population size is then 
4.6.

What should the correct design setup in STATA be?

Thanks for your help.

Susan
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/