Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Usability of Adaptive Cluster Sampling for Identifying the Land Producing ‘hybrid Borocrop’ in Bangladesh

Saiful Alam Chowdhury
Institute of Statistical Research and Training (ISRT) University of Dhaka, Dhaka - 1000, Bangladesh

Syed Shahadat Hossain
Institute of Statistical Research and Training (ISRT) University of Dhaka, Dhaka - 1000, Bangladesh

Abstract/ Executive Summary:

The coverage of Hybrid crops usage in Bangladesh is rare as well as of clustered characteristics. So the applicability of the adaptive cluster sampling is planned to be visualized by a simulation consisting re-sampling of households from the Agriculture census (2008) data collected by Bangladesh Bureau of Statistics. While 2016 Agriculture census is a continuous process, a demonstration of the applicability of such a design may help planners to set preparation of sampling frame such a way that future usage of these designs can be enabled. The specific objectives of this study included to find estimates of the proportion of household cultivating Hybrid Boro using simple random sampling and adaptive cluster sampling methods and to compare them in respect of Monte Carlo standard errors, ultimate sample sizes and ability of capturing rare characteristic. Different initial sample sizes are considered for this comparison. In terms of Monte Carlo standard error of the proportion of HH cultivating Hybrid Boro, the adaptive cluster sampling was found to perform similar to the simple random sampling whereas the adaptive cluster sampling method has shown better performance in capturing units with characteristic of interest than that by simple random sampling method. The average ultimate sample size realized by the application of the adaptive cluster sampling were found to be moderate and is about 10 to 20 p.

Keywords: Conventional sampling, Adaptive Cluster Sampling, Hybrid Boro, Agriculture census.

Location:

Start Date:

End Date:

Program Area: Socio-economic and Policy

Non-commodity: Performance

Objectives:

The main objective of the study is to investigate whether the adaptive methods can be more advantageously utilized for estimating the proportion of land producing hybrid crops. The method is compared with the conventional simple random sampling in terms of standard error and coverage strength to see whether adaptive cluster sampling has any advantage over the conventional method.

Methodology:

ADAPTIVE CLUSTER SAMPLING Adaptive cluster sampling refers to design in which an initial set of units is selected by some probability sampling procedure, and, whenever the variable of interest of a selected unit satisfies a given criterion; additional units in the neighborhood of that unit are added to the sample. Thompson (1990) developed an adaptive cluster sampling (ACS) procedure. ACS has a wide range of application in ecological and environmental studies when the population has a rare and cluster tendency. the neighborhood units are also included in the sample and this process is terminated when a full cluster of unit is selected. The ACS method, although was originally designed for geographical or spatial sampling, the method has the potential to be utilized for house hold sampling as well. The methodology adjusted for sampling of HH for estimating Hybrid Boro usage in Bangladesh is described in this section. This description has also taken into consideration the stratification for division. Other topics are as follows: Estimator of the Population Proportion, A COMPUTER SIMULATION, Agriculture census Data 2008, Simulation, Monte Carlo Mean and Standard Deviation, RESULTS AND DISCUSSION, CONCLUSION AND POLICY RECOMMENDATIONS.

Source of Information: Bangladesh J. Agric. Econs. XXXVIII, 1&2 (2016-2017) 103-113

Article Link (Online Journal):

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

The potential risk of this scenario could lead to a more expensive survey by accumulating a large sample size. For the estimation of proportion of HH cultivating Hybrid Boro, the simulation study has revealed that the ultimate sample sizes are not too large in respect of the initial sample sizes for each of the divisions and each of the sample sizes. The ultimate sample sizes for Barisal and Sylhet division are found to be increased less than that of other divisions. This reveals the fact that rarer the characteristics of interest, smaller the increase in ultimate sample sizes. Yet again, the minimal increase in the sample size to constitute an ultimate sample size may be traded off with the minimization of risk of ending up with very small number of units containing the targeted characteristics. The results, in a summary, reflects possible use of the ACS design for future studies in Bangladesh where a quick estimation of a rare agricultural HH characteristic is of interest. It is, therefore, required to be kept into planning of sampling frame and Integrated multipurpose sampling (IMPS) preparation of the country.

Knowledge Management: Journal