DATA INTEGRATION NONPROBABILTY SAMPLES
DATA INTEGRATION NONPROBABILTY SAMPLES
Survey data integration refers to the combination of data from different surveys or sources.
Traditional probability sample surveys suffer from low response rates and can be prohibitively expensive. New data sources, including volunteer web surveys and big data (e.g., social media, Google Trends, sensors, etc.), are more convenient, granular, and timely, but are also subject to various biases.
How can we integrate multiple data sources to improve statistical inference and better understand social phenomena?
Check my publication and coference presentation list for more insight on my work!
Moretti, A., Salvatore, C., and Kraemer, F. (2026). A Data Quality Framework for Integrated Data via Statistical Matching, International Journal of Population Data Science, 11(5), https://doi.org/10.23889/ijpds.v11i5.3694
Salvatore C., (2024). Inference with non-probability samples and survey data integration: a science mapping study. Metron, https://doi.org/10.1007/s40300-023-00243-6
Salvatore C., Biffignandi S., Sakshaug J., Wiśniowski A., Struminskaya B., (2023). Bayesian integration of probability and non-probability samples for logistic regression. Journal of Survey Statistics and Methodology, https://doi.org/10.1093/jssam/smad041 - Supplemented with a Shiny App
Salvatore C., Biffignandi S., Bianchi A., (2023). Augmenting Business Statistics Information by Combining Traditional Data with Textual Data: A Composite Indicator Approach, Metron