Development of Alternative Sampling Design and Survey Strategies for the Conduct of the CAF

 

Background and Rationale

The Development of Alternative Sampling Design and Survey Strategies for the Conduct of the Census of Agriculture and Fisheries (CAF) is one of the activities in the preparation of the proposal on Ensuring Quality Statistics from the 2002 Census of Agriculture and Fisheries and Statistics through Research and Training. This project is a collaborative effort of the Statistical Research and Training Center (SRTC), together with the National Statistics Office (NSO) and Bureau of Agricultural Statistics (BAS). The aim is to come up with an optimum design and strategy for the conduct of the said census on 2012, given the available resources of the Philippine Statistical System and constraints inherent in its situation and conditions. The project was conceived with an urgent need to evaluate the efficiency of the 2002 CAF way back in 2004.

Expected Outputs/Scope of Work

The project consisted of two components: the List Frame and the Area Frame. The compositions of the two were dependent on specific project activities undertaken at specific stage or phase. The List Frame focused on the evaluation of the sampling design and strategies utilized in the previous Census of Agriculture (CA) and on the identification of discrepancies in the said survey. Along this line, a review of the strengths and weaknesses of the past agricultural censuses conducted in the country and those in other countries was made. Two options on the adoption of a modular approach on the sampling design for CA 2012 were proposed by the team, depending on availability of financial resources: (1) Complete enumeration of the 50% sampled barangay with municipality as domain and barangay as the Primary Sampling Units (PSUs); and (2) Adoption of the 1991 CA sampling design using certainty and non-certainty barangays with municipality as domain and barangay as the PSUs using proportion of agricultural households based on the data that will be collected from the 2010 Census of Population and Housing (CPH) as the stratification variable . However, a need to do a careful study on these options would still have to undertaken in the Phase II of List Frame Project. For the 2010 Census of Fisheries (CF), there is a need to undertake similar studies/researches on the past fisheries censuses in order to come up with an appropriate sampling design and survey approaches for the said census. Moreover, the feasibility of simultaneous conduct of the CA and the CF will be looked into on the phase of the project. Further review/study of questions to be included in the agriculture and fishery censuses will be given more priority as per project workplan.

The creation of Area Frame is a scheme with an end view of considering the “area sampling frame for agriculture” as possible option to be tested to compliment the current list frame to be used in the conduct of CA 2012. Among the identified uses of area frame technologies were multi-purpose information system like land use, land cover environment-related information, and photographing of landscape; health related application; and enhancement of GIS technology. Trainings on the applications of GIS/GPS and remote sensing in the construction of area frame were one of the activities done in this project in preparation of the ground truthing activity conducted in Quirino, Isabela - the area of study.

Activities such as stratification of PSUs, breaking PSUs into Secondary Sampling Units (SSUs) with identifiable boundaries, selection and interview of sample farmers, and simulation of sampling design estimation procedures will be done as part of Phase II of Area Frame Project. That is, pilot testing of area frame based agricultural census will be conducted to provide more comprehensive information on the advantages or disadvantages of using area frame sampling in conducting a census and also to present an alternative design that is not only cheaper but will give more accurate and reliable estimates. The project on Area Frame is also geared towards the development of an actual area sampling design since sampling procedure methodology was only pre-tested in Quirino, Isabela – the area of study.