64th ISI World Statistics Congress - Ottawa, Canada

64th ISI World Statistics Congress - Ottawa, Canada

Redesigning of the Commercial Livestock and Poultry Survey in the Philippines


Quindale Caraos



64th ISI World Statistics Congress - Ottawa, Canada

Format: CPS Paper

Keywords: agricultural, sampling, survey

Session: CPS 80 - Survey statistics V

Thursday 20 July 8:30 a.m. - 9:40 a.m. (Canada/Eastern)


The Commercial Livestock and Poultry Survey (CLPS) is one of the major agricultural surveys conducted by the Philippine Statistics Authority (PSA). This survey aims to generate estimates on supply, disposition, average farmgate price, and other related data of livestock and poultry commodities from establishments. The old design of CLPS uses stratified random sampling with maximum farm/housing capacity as the stratification variable. In the old design, the number of strata vary per animal type per province which imposes difficulty in terms of field operations and since the stratum boundaries are not updated regularly, there is a possibility that some farms/establishments that have increased in terms of maximum farm/housing capacity are still on lower strata. Given the situation, a new sampling design is proposed in order to generate reliable and accurate estimates at the provincial level with the use of the regularly conducted Updated List of Establishments (ULE) as its new sampling frame.
After performing sampling experiments and simulations, the new sampling design for CLPS is a one-stage Stratified-Probability Proportional to Size (PPS) where the domain is the province, and the establishment is the primary sampling unit (PSU). The animal type serves as a stratification variable and the size measure is the maximum farm/housing capacity of the establishment. The new sampling design has better design effect (DEFF) and coefficient of variations (CVs) for indicators estimated in CLPS as compared to the old design.