Survey researchers and practitioners have always combined data from various data sources (auxiliary data from Censuses, administrative registers, GIS data) to enhance sampling designs and estimation: from stratification and probability proportional to size sampling to balanced sampling designs (see e.g. [9], for a review), from ratio estimation to modelassisted survey estimation with modern regression techniques [1], from post-stratification to calibration and its extensions [3, 7]. Statistical data integration represents the new frontier of combining information from representative samples with data from multiple sources some of which may have uncontrolled selection mechanisms (non-probability samples, big data, web-scraped data, integrated register data). This special issue contains a first selection of the papers presented at the Seventh Italian Conference on Survey Methodology, ITACOSM2022, held in Perugia in June 2022 with a focus on “Survey Methods for Statistical Data Integration and New Data Sources”. ITACOSM is a bi-annual international conference organized by the Survey Sampling Group of the Italian Statistical Society whose aim is promoting the scientific discussion on the theoretical and applied developments of survey sampling methodologies in the fields of economics, social and demographic sciences, official statistics and environmental sciences. In particular, ITACOSM2022 has provided a showcase for methods and applications that combine sources of data for better sampling strategies using classical survey methods or new data science tools such as Machine Learning methods.

Foreword to the special issue on “Survey Methods for Statistical Data Integration and New Data Sources”

Bertarelli, Gaia;
2023-01-01

Abstract

Survey researchers and practitioners have always combined data from various data sources (auxiliary data from Censuses, administrative registers, GIS data) to enhance sampling designs and estimation: from stratification and probability proportional to size sampling to balanced sampling designs (see e.g. [9], for a review), from ratio estimation to modelassisted survey estimation with modern regression techniques [1], from post-stratification to calibration and its extensions [3, 7]. Statistical data integration represents the new frontier of combining information from representative samples with data from multiple sources some of which may have uncontrolled selection mechanisms (non-probability samples, big data, web-scraped data, integrated register data). This special issue contains a first selection of the papers presented at the Seventh Italian Conference on Survey Methodology, ITACOSM2022, held in Perugia in June 2022 with a focus on “Survey Methods for Statistical Data Integration and New Data Sources”. ITACOSM is a bi-annual international conference organized by the Survey Sampling Group of the Italian Statistical Society whose aim is promoting the scientific discussion on the theoretical and applied developments of survey sampling methodologies in the fields of economics, social and demographic sciences, official statistics and environmental sciences. In particular, ITACOSM2022 has provided a showcase for methods and applications that combine sources of data for better sampling strategies using classical survey methods or new data science tools such as Machine Learning methods.
2023
81
File in questo prodotto:
File Dimensione Formato  
s40300-023-00248-1.pdf

embargo fino al 01/08/2024

Tipologia: Documento in Post-print
Licenza: Accesso gratuito (solo visione)
Dimensione 162.07 kB
Formato Adobe PDF
162.07 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/5023304
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 2
social impact