ESRA 2027 sessions by theme
Back to Overview of Sessions
Advances in spatial sampling methods and their applications |
| Coordinator 1 | Professor Alina Matei (University of Neuchatel, Switzerland) |
| Coordinator 2 | Mr Thomas Merly-Alpa (INSEE, France) |
Session Details
Over the last decade, the use of georeferenced data, such as geospatial and satellite data, has grown significantly. A variety of spatial sampling methods have been developed for use in different fields, including environmental science, social sciences, and official statistics. In this context, two important topics are:
a) Spatially balanced sampling, which aims to provide the best possible coverage of a given area while avoiding concentration of sampled units, has attracted considerable interest. It offers great flexibility of application and can extend beyond geographic space to include auxiliary variables. These methods improve the accuracy of the estimators, especially if there is spatial autocorrelation of the variable of interest, without requiring an increase in the sample size.
b) Combining non-probability sample data with probability sample data within a spatial framework. Although probability samples are regarded as the gold standard for population-based research, non-probability samples are frequently used in different fields due to their low cost, time efficiency, and lack of sampling frame information. Recently, data integration approaches incorporating mass imputation, propensity score weighting and calibration have been used to improve the representativeness of non-probability samples.
Possible topics for the session include:
1) new methods and/or applications related to spatially balanced sampling,
2) new methods and/or applications of combining non-probability and probability sample data in a spatial context,
3) new methods and/or applications involving sampling and remote sensing (e.g. satellite images, aerial cameras, scanned maps).
We welcome contributions from all fields and encourage submissions that present novel methods, ideas, applications and solutions to methodological problems involving spatial sampling and georeferenced data.
The session is proposed on behalf of the `Surveys' group of the French Statistical Society (SFdS).
