To provide reliable pre-election forecasts it is necessary to estimate the election results as precisely as possible in individual electoral districts. Such estimates are traditionally based on a series of pre-election surveys in the individual electoral districts. The article presents an alternative method of making pre-election estimates, which combines the results of national representative surveys and information on the spatial distribution of party support from the previous election, and the results of the model-based estimates are compared with traditional survey-based estimates. The article demonstrates the usefulness of the model-based approach and discusses the conditions that either support or impede the quality of model-based estimates.
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.