SSSI is a professional spatial sampling and statistical inference tool. It can be used for sampling and statistical inference in environment, resources, land, ecological, social and economic sciences.
On a planned monitoring network (such as a planned sampling project in agriculture, demography, economy, environment, or epidemics, for example):
- Calculation of the optimum distribution and density of sample sites to form a highly efficient spatial sampling project or monitoring network;
On an existing monitoring network (such as an existing weather observation network, an existing epidemic surveillance network, an existing satellite monitoring scheme for example):
- Recommend the best overall valuation methods and recommendations to improve the monitoring network (based on monitoring the target characteristics and distribution of observations);
On published statistics (such as regional greenhouse gas (GHG) emissions; prevalence of a disease in a region, areas of contaminated soil in a region)
- Evaluation of the statistical errors (through the study of its sample distribution, density, statistical methods).
The software provides a total of six sampling and estimation methods: simple random sampling, systematic sampling, stratified random sampling, spatial random sampling, spatial stratified sampling, and sandwich estimation. In this software, we implement a new estimation method- the "sandwich" estimator, which is one of the major features of this software with a higher efficiency for sampling and statistical inference. On the basis of spatial stratified sampling, we develop a reporting layer, composed of the final reporting units that the user wishes to use, for example: county and/or provincial boundaries, watersheds, an artificial grid.
1. Wang JF, Haining R, Liu TJ, Li LF, Jiang CS. 2013. Sandwich spatial estimation for multi-unit reporting on a stratified heterogeneous surface. Environment and Planning A, 45, 2515–2534.Download
2. Wang, JF, Jiang, CS, Hu, MG, Cao, ZD, Guo, YS, Li, LF, Liu, TJ, Meng, B. 2012. Design-based spatial sampling: Theory and implementation. Environmental Modelling & Software, 40, 280-288.Download
3. Wang JF, Haining R, Cao ZD. 2010. Sample surveying to estimate the mean of a heterogeneous surface: reducing the error variance through zoning. International Journal of Geographical Information Science, 24, 523-543. Download
4. Wang JF, Jiang CS, Li LF, Hu MG. 2009. Spatial Sampling and Statistical Inference (in Chinese). Beijing: Science Press. Browse
5. Wang JF, Liu JY, Zhuang DF, Li LF & Ge Y. 2002. Spatial sampling design for monitoring the area of cultivated land. International Journal of Remote Sensing, 13, 263-284.
Acknowledgement: spatial sampling survey methods and the theory and its software (SSSI) are funded by the National Natural Science Foundation (No. 40471111) and National 863 hi-tech project (No. 2006AA12Z215).
This software is composed of SuperMap GIS and sampling systems. SuperMap GIS is responsible for the graphic view, data management and map browser. The sampling system is responsible for reading and writing sampling project files, setting the sampling region, parameters, generating sample sites and presenting statistical results.
The sampling process in the software is divided into three stages. The first stage is to calculate a sample size to achieve a target level of accuracy. The second stage is to set up sample sites and obtain sample values. The third stage is to report the results of statistical inference. In the existing sampling theory, the method of calculation of sample size, sample site settings and estimation of the population uses the same sampling model. In this software, the calculation of sample size, sample site settings and the population estimation can be chosen by different models, the strategy may result in a higher sampling efficiency and a higher posterior precision of an estimation.