This thesis analyzes the socio-demographic factors that influence voter turnout in Italian municipalities, with a particular focus on the political and European elections from 2013 to 2024. Using a large dataset aggregated at the municipal level, correlations between variables, geographic visualizations and dimensionality reduction techniques are explored. The analysis continues with the construction and comparison of different statistical models in order to identify the most relevant variables. The presence of spatial correlation is discussed and a geostatistical model is proposed to assess its residual impact. Finally, the advantages and inferential limits of the ecological approach adopted are reflected upon.
Questa tesi analizza i fattori socio-demografici che influenzano l'affluenza elettorale nei comuni italiani, con particolare attenzione alle elezioni politiche ed europee dal 2013 al 2024. Attraverso l'utilizzo di un ampio dataset aggregato a livello comunale, vengono esplorate correlazioni tra variabili, visualizzazioni geografiche e tecniche di riduzione della dimensionalità. L'analisi prosegue con la costruzione e il confronto di diversi modelli statistici al fine di individuare le variabili più rilevanti. Si discute la presenza di correlazione spaziale e si propone un modello geostatistico per valutarne l'impatto residuo. Infine, si riflette sui vantaggi e sui limiti inferenziali dell'approccio ecologico adottato.
Partecipazione elettorale in Italia: un'analisi ecologica a livello comunale
MENEGON, NICOLA
2024/2025
Abstract
This thesis analyzes the socio-demographic factors that influence voter turnout in Italian municipalities, with a particular focus on the political and European elections from 2013 to 2024. Using a large dataset aggregated at the municipal level, correlations between variables, geographic visualizations and dimensionality reduction techniques are explored. The analysis continues with the construction and comparison of different statistical models in order to identify the most relevant variables. The presence of spatial correlation is discussed and a geostatistical model is proposed to assess its residual impact. Finally, the advantages and inferential limits of the ecological approach adopted are reflected upon.| File | Dimensione | Formato | |
|---|---|---|---|
|
Menegon_Nicola.pdf
accesso aperto
Dimensione
1.19 MB
Formato
Adobe PDF
|
1.19 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/92963