Page 72 - Contributed Paper Session (CPS) - Volume 1

CPS1071 Ibarra Aaron R. Poliquit et al.
of animals on the milk line is the size measure variable. Additionally, for both
semesters of Misamis Oriental, MAPE of estimates exceeded 100% when the
number of female breeders is the size measure variable. The results are varied
for the provinces of Batangas, Bohol and Laguna. These results are supported
by the fact that the number of animals on the milk line has the highest
correlation to total milk production among all possible size measure variables.
However, these results might be inconclusive since total sample size is larger
when the number of animals on the milk line is the size measure variable than
when number of female breeders.

4. Discussion and Conclusion
The objective of computing barangay sample size using LH-algorithm was
achieved. However, a stratified design using LH-algorithm was found to be
inappropriate since for minor-producing provinces, some computed stratum
sample sizes were equal to one which makes variance estimation impossible.
A PPS systematic sample of barangays with number of animals on the milk line
as size measure and a dairy farm sample size of 25 per sampled barangay,
generate CV and APE below 10% for major-producing provinces (i.e.,
provinces that have barangays with 25 or more dairy farms).
Simulation results provide an evidence that using the number of animals
on the milk line as size measure may be better than using the number of
female breeders as size measure in terms of reliability of provincial milk
production estimates.

References
1. Compilation of Sampling Methodologies used in Agriculture and Fisheries
Survey (2015). Survey Designs Research and Development Section,
Philippine Statistics Authority.
2. Santos, K.C. (2014). Improvement of Methodology in Semi-Annual Survey
of Dairy Enterprises (SSDE) Phase I. University of the Philippines, Diliman,
Quezon City.

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