Method of Area Frame Sampling Using Probability Proportional to Size Sampling Technique for Crops’ Surveys: A Case Study in Pakistan

The Crops’ estimates have been greatly concerned by the Government of Punjab (Pakistan) all the times. Crop Reporting Service (CRS), Agriculture Department, Punjab, as a unique and the largest statistical organization in Punjab, has been working on agricultural statistics using the sampling technique, List Frame Sampling (LFS), for conducting surveys to gather information regarding crops acreage, cost of production, crops yield and other agricultural items since 1978. The development of the rural economy in Pakistan brings new problems and challenges to the methods of agricultural statistics. The back bone of agricultural statistics is the sampling technique, LFS, in which primary sampling unit is a village and an enumerator has to survey the whole village whatever the size of the village causing an increase in non-sampling error. In spite of the sufficient area coverage, representation of population in the sample is not satisfactory. The solution is Area Frame Sampling Technique in which primary sampling unit is a Segment of a specific acreage covering maximum dimensions of the population land. In this paper a method of Area Frame Sampling (AFS) has been proposed. As most of the research papers focus on the method of AFS through Geographical Information System (GIS) technique. But in this paper two-stage statistical sampling technique has been used to achieve the same objective in an efficient and economical way. In the first stage Probability Proportional to Size Sampling (PPS) has been used. Here size is cropped area of a Union Council (UC), the smallest geographical cluster in Punjab, Pakistan. In the second stage Simple Random Sampling (SRS) has been used. Here Primary Sampling Units are Segments of land of a village. The results show that the problem of non-representation of agricultural land is minimized and, consequently, getting better estimates in terms of precision using less amount of land data. It is recommended that this method can be extended for multiple stages of sampling and for multiple measures of sizes.

Linking Synthetic Populations to Household Geolocations: A Demonstration in Namibia

Whether evaluating gridded population dataset estimates (e.g., WorldPop, LandScan) or household survey sample designs, a population census linked to residential locations are needed. Geolocated census microdata data, however, are almost never available and are thus best simulated. In this paper, we simulate a close-to-reality population of individuals nested in households geolocated to realistic building locations. Using the R simPop package and ArcGIS, multiple realizations of a geolocated synthetic population are derived from the Namibia 2011 census 20% microdata sample, Namibia census enumeration area boundaries, Namibia 2013 Demographic and Health Survey (DHS), and dozens of spatial covariates derived from publicly available datasets. Realistic household latitude-longitude coordinates are manually generated based on public satellite imagery. Simulated households are linked to latitude-longitude coordinates by identifying distinct household types with multivariate k-means analysis and modelling a probability surface for each household type using Random Forest machine learning methods. We simulate five realizations of a synthetic population in Namibia’s Oshikoto region, including demographic, socioeconomic, and outcome characteristics at the level of household, woman, and child. Comparison of variables in the synthetic population were made with 2011 census 20% sample and 2013 DHS data by primary sampling unit/enumeration area. We found that synthetic population variable distributions matched observed observations and followed expected spatial patterns. We outline a novel process to simulate a close-to-reality microdata census geolocated to realistic building locations in a low- or middle-income country setting to support spatial demographic research and survey methodological development while avoiding disclosure risk of individuals.

Linking Synthetic Populations to Household Geolocations: A Demonstration in Namibia

Whether evaluating gridded population dataset estimates (e.g. WorldPop, LandScan) or household survey sample designs, a population census linked to residential locations are needed. Geolocated census microdata data, however, are almost never available and are thus best simulated. In this paper, we simulate a close-to-reality population of individuals nested in households geolocated to realistic building locations. Using the R simPop package and ArcGIS, multiple realizations of a geolocated synthetic population are derived from the Namibia 2011 census 20% microdata sample, Namibia census enumeration area boundaries, Namibia 2013 Demographic and Health Survey (DHS), and dozens of publicly available spatial datasets. Realistic household latitude-longitude coordinates are manually generated based on public satellite imagery. Simulated households are linked to latitude-longitude coordinates by identifying distinct household types with multivariate kmeans analysis, and modelling a probability surface for each household type using Random Forest machine learning methods. We simulate five realizations of a synthetic population in Namibia's Oshikoto region, including demographic, socioeconomic and outcome characteristics at the level of household, woman, and child. Comparison of variables in the synthetic population were made with 2011 census 20% sample and 2013 DHS data by primary sampling unit/enumeration area. We found that synthetic population variable distributions matched observed observations and followed expected spatial patterns. We outline a novel process to simulate a close-to-reality microdata census geolocated to realistic building locations in a low- or middle-income country setting to support spatial demographic research and survey methodological development while avoiding disclosure risk of individuals.

Weighting Procedure of the Austrian Microcensus using Administrative Data

The Austrian microcensus is the biggest sample survey of the Austrian population, itis a regionally stratied cluster sample with a rotational pattern. The sampling fractionsdier signicantly between the regions, therefore the sample size of the regions is quitehomogeneous. The primary sampling unit is the household, within each household allpersons are surveyed. The design weights are the input for the calibration on populationcounts and household forecasts. It is performed by iterative proportional tting. Untilthe third quarter of 2014 only demographic, regional and household information wereused in the weighting procedure. From the fourth quarter 2014 onwards the weightingprocess was improved by adding an additional dimension to the calibration, namely alabour status generated from administrative data and available for the whole population.Apart from that some further minor changes were introduced. This paper describes themethodological and practical issues of the microcensus weighting process and the varianceestimation applied from 2015 onwards. The new procedure was used for the rst timefor the forth quarter of 2014, published at the end of March 2015. At the same time, allprevious microcensus surveys back to 2004 were reweighted according to the new approach.

A serological prevalence survey of Brucella abortus in cattle of rural communities in the province of KwaZulu-Natal, South Africa : article

A serological survey of Brucella abortus in cattle originating from communal grazing areas of Kwa Zulu Natal was carried out between March 2001 and December 2003. The survey was designed as a 2-stage survey, considering the diptank as the primary sampling unit. In total 46 025 animals from 446 diptanks of 33 magisterial districts were sampled and tested using the Rose Bengal test and Complement Fixation Test. The apparent prevalence at district level was adjusted for clustering, diagnostic test sensitivity and specificity, and mapped using ArcView version 3.3. The prevalence of brucellosis in communal grazing areas of Kwa-Zulu Natal was found to be 1.45 % (0.84-2.21 %) and varied from 0 to 15.6% between magisterial districts. In 19 of the 33 magisterial districts no serological reactors were observed. A large variation in prevalence was found within diptank areas. Brucellosis was found to be most prevalent in the northeastern area of the province. The findings of the survey are discussed.

DETERMINANTS OF CURRENT CONTRACEPTIVE USE AND METHOD CHOICE IN MONGOLIA

SummaryThis study examines the determinants of current contraceptive use and method choice in Mongolia using data from the 1998 Mongolian Reproductive Health Survey and 2000 Mongolian Population and Housing Census. Since 1976, access to modern contraceptives has been liberalized and all restrictions on the use, distribution and import of contraceptives were removed in 1989. There were some increases in the use of modern contraceptives among married women in the 1990s; however, at the start of the twenty-first century the IUD and periodic abstinence remain the most widely used methods. Women with higher levels of education are more likely to be current users of contraception, and if they are current users, they are more likely to choose the IUD and traditional methods. Women living in rural areas have a higher probability of using contraception and are more likely to choose the IUD and traditional methods. Significant variations exist between primary sampling units in current contraceptive use and in the choice of modern methods. Community-level variables were important predictors in reducing variation between primary sampling unit, when other modern methods were compared with traditional methods.