Page 47 - Contributed Paper Session (CPS) - Volume 5

CPS861 Madeline Dumaua C.
Comparative analysis of design-based and
synthetic estimates of ownership of farm
equipment and facilities for the 2002 Census of
Agriculture
Madeline Dumaua Cabauatan
Asian Development Bank, Mandaluyong City, Philippines, line.dumaua@gmail.com

Abstract
This study aims to analyze the feasibility of using synthetic estimator in
estimating the total number of farm equipment and facilities for the 2002
Census of Agriculture in the municipal level. The estimation procedures
employed were design-based and synthetic estimation. The data used came
from the 1991 Census of Agriculture and Fisheries and the 1990 Census of
Population and Housing. The variables total number of plows owned, total
number of harrows owned, total number of sprayers owned, total number of
threshers owned, and total number of handtractors owned were considered in
the study. The resulting estimates were evaluated and compared based on the
ratio of the absolute bias to the standard deviation, Relative Root Mean Square
Error (RRMSE), and Coefficient of Variation (CV). In synthetic estimation, the
total number of households was used as an auxiliary variable. Compared to
synthetic estimation, estimates generated using the design-based estimator
were more accurate and precise; more reliable; and the biases of the estimates
were negligible. Synthetic estimator could have work under the assumption
that the municipality has the same characteristics as the province but the study
has been limited to using one auxiliary variable only.

Keywords
auxiliary variable, design-based estimation, farm equipment and facilities,
synthetic estimation

1. Introduction
a. Background of the Study
There is a higher demand nowadays on estimates of various characteristics
as more quantitative information required for decision making in government
and industry arises. These estimates are often generated from nationwide
sample surveys. However, estimation from nationwide survey data has
limitations in terms of producing reliable microdata since production of small
area statistics has emerged as a pressing and frequently difficult and costly
problem.
The Census of Agriculture and Fisheries (CAF) is essential to our
development since agriculture remains to be dominant sector of the economy
of the Philippines. CAF is a complete count of all farms and fisheries in the

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