In this paper, we consider the meta-analysis approach as a special case of multilevel modelling in order to identify the potential sources of dissimilarities among pre-election poll accuracy measures. The predictive accuracy measure is computed to compare the pre-electoral poll result to the actual result. It is introduced in a hierarchical model as a dependent variable and the poll characteristics as explanatory variables. In the model each outcome is affected by a specific sampling error assumed to have a normal distribution and a known variance. Employing such a multilevel approach, we conduct a meta-analysis of pre-election poll accuracy measures in Parliamentary elections in Italy from 2001 to 2008.
|Titolo:||Multilevel models in meta-analysis of pre-election poll accuracy measures|
|Autori interni:||D'AGATA, ROSARIO GIUSEPPE|
|Data di pubblicazione:||2013|
|Appare nelle tipologie:||1.1 Articolo in rivista|