Page 68 - Contributed Paper Session (CPS) - Volume 1

CPS1071 Ibarra Aaron R. Poliquit et al.
”take none” stratum and no sample barangay were drawn from this group. A
5
simulation using R for different scenarios were explored to determine the
number of sample barangays for both animals on the milk line and female
breeders as stratification variables using different combinations of number of
strata (3, 4 and 5 strata) and CVs (1.0%, 5.0% and 8.0%) per province.

2.2 Probability Proportional to Size (PPS) Systematic
Treating the stratification variable as size measure, a PPS systematic
sample of barangays were composed mostly of barangays with large size
measures and fewer number of barangays with small size measures. This
means that the sample barangays selected via PPS systematic sampling were
similar to the sample barangays selected via stratified design with ”take none”
and ”take all” strata. So, PPS systematic was used as the 1st stage sampling
6
method for the selection of sample barangays .

2.3 1 and 2 Stage Sample Selection Scheme
nd
st
The secondary sampling unit (SSU) for this study is the dairy farm or dairy
st
nd
enterprise. The steps for the 1 and 2 stage sample selection as follows:
1. Compute = 2 × where is the sample size computed such
0
0
that the stratum number is 5, and the CV is 0.05 for a specific
stratification variable, while the constant 2 is the design effect
adjustment due to cluster sampling.
2. Compute the PPS Systematic sampling of barangays.

3. Label a barangays as the ℎ sampled barangay when its
corresponding cumulative size measure is less than or equal to +
( + 1) × Replace the sample duplicates by doing steps 2 and 3 for
the unselected barangays and by setting sample size equal to the
number of duplicates.

4. Repeat step 4 until all barangays in the sample are distinct.
5. Divide the sample into 2, one for each semester, such that each
division inherits the characteristics of a sample taken using PPS
Systematic.

6. In each sampled barangay, select dairy farms via Simple Random
Sampling Without Replacement (SRSWOR).

7. Explore different dairy farm sample sizes (i.e., m=5, 10, m=15, m=20,
and m=25) in each sampled barangay.

57 | I S I W S C 2 0 1 9