Exploration and description /Estimation
Putting the results for the two scenarios together we get:
|
Dip-tank areas
|
Population estimates
|
Sub-region
|
Population
|
Sample
|
Scenario 1
|
Scenario 2
|
Lobamba
|
13
|
4
|
24,403 ± 4,276
|
20,023 ± 5,327
|
Malandzela
|
12
|
7
|
24,884 ± 4,310
|
22,408 ± 3,324
|
Mayiwane
|
20
|
15
|
77,510 ± 6,960
|
66,038 ± 6,680
|
Mbabane
|
28
|
20
|
84,489 ± 9,996
|
73,841 ± 6,883
|
Ntfonjeni
|
13
|
7
|
32,897 ± 4,421
|
23,462 ± 3,347
|
Piggs-Peak
|
14
|
5
|
46,526 ± 6,555
|
27,008 ± 4,481
|
Total
|
100
|
58
|
290,709 ± 15,740
|
232,780 ± 12,756
|
As all sub-regions are sampled there is no further sampling variation to add when calculating the standard error for the grand total. There is, therefore, no additional term to add to the formula for the sampling variance.
The standard error for the total population is thus the square root of the sum of variances for each sub-region. For scenario 1 this is:
√(4,2762 + 4,3102 + . . . + 6,5552) = 15,740.
Similarly the standard error for scenario 2 is 12,756.
|