Page 48 - Contributed Paper Session (CPS) - Volume 5

CPS861 Madeline Dumaua C.
Philippines. The total number of farm equipment and facilities are few of the
statistics generated by the National Statistics Office (NSO) through the Census
of Agriculture. These statistics however, provide estimates only at the national,
regional and provincial levels. Equipment and facilities refer to farm
equipment and facilities used for agricultural activities during the reference
period January 1, 1991 to December 31, 1991. Equipment and facilities that
were unusable or beyond repair during the reference period were excluded.
Due to great cost allotted to this census, it is being conducted once every
ten years. In the 1960, 1971, and 1980 census rounds, all barangays in the
country were completely enumerated and sampling of farm households was
done in each barangay. Complete enumeration of farm households have not
been possible in the 1991 and 2002 census rounds since the government
trimmed down the national budget in favor of the 1990 Census of Population
and Housing (CPH). Due to less support for CAF than CPH in general, the need
for extensive use of sampling to reduce costs is more acute in the former.
The 1991 CAF design was a systematic selection of 50% of the barangays
in each municipality ordered with respect to total farm area and the selection
with certainty of the barangay with the largest farm area based on the 1980
CAF. In each sampled barangay, all farm households were completely
enumerated. Moreover, complete enumerations of all farm household were
conducted of four provinces namely: Bukidnon, Marinduque, Isabela, and
Laguna to validate the results of the study. Results from these provinces
yielded precise estimates for most statistics at the provincial level but barely
enough at the municipal levels.
The same design was implemented in the 2002 CAF except that 25% of the
barangays in each municipality ordered with respect to total farm area and the
selection with certainty of the barangay with the largest farm area based on
the 1980 CAF are included. In each sampled barangay, all farm households
were completely enumerated. Nevertheless, this accounted for a very small
sample in the area leading to a lesser precision of estimates at all small areas.

b. Statement of the Problem
According to Morales as cited by Ghosh and Rao (1994), for the past few
decades, sample surveys have taken the place of complete enumeration or
census as a more cost-effective means of obtaining information on wide-
ranging topics of interest at frequent intervals over time. Sample survey data
certainly can be used to derive reliable estimates for totals and means for large
areas or domains. However, the usual direct survey estimators for small areas
based only on the data from the sampled units in the larger area, are likely to
yield unacceptably large standard errors due to excessively small size of the
sample in the area. With that regard, the field of small area statistics is given
more focus on estimating characteristics at small levels.

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