Estimating the sources of bias : a predictive model to analyze polling performance for the Congressional race in Mexico
It proposes a Montecarlo simulation analysis to estimate the sources of bias in the main national surveys for the Congressional Race in Mexico from 1997 to 2012. According to similar approaches for other electoral systems, the main sources of error analyzed were the pollster bias. The results of suc...
| Autor principal: | Torhton Granados, Miguel Angel |
|---|---|
| Outros Autores: | Tribunal Superior Eleitoral |
| Tipo de documento: | Artigo |
| Idioma: | English |
| Publicado em: |
2020
|
| Assuntos: | |
| Obter o texto integral: |
|
| Resumo: |
It proposes a Montecarlo simulation analysis to estimate the sources of bias in the main national surveys for the Congressional Race in Mexico from 1997 to 2012. According to similar approaches for other electoral systems, the main sources of error analyzed were the pollster bias. The results of such analysis may be indicative of some of the main problems facing electoral polls in Mexico. The direction and patterns of bias for each pollster were identified. The sources and the characteristics of systematic errors could be of great value to understand the complexities of the Mexican electoral system and the challenges for polling agencies. |
|---|