Challenging the Manifesto Project data monopoly: Estimating policy position time-series using expert and mass survey data

bruinsmagemenis2017.pdf565 KB

Abstract:

Whenever researchers need to test theories and hypotheses using longitudinal data of political parties’ ideological and policy placement, they have little choice. Researchers are often constrained to use the Manifesto Project data, despite the extensive evidence that has challenged its reliability and validity. In this paper we show that it is possible to construct a unique and rich time-series of policy placements by combining expert and mass survey data, and addressing the problem of missing values through the Amelia II multiple imputation algorithm. Using data from Germany, the Netherlands, and Greece, we estimate the positions of parties on the left-right dimension and on a two-dimensional (socio-economic and socio-cultural) space, and show how the estimates outperform the Manifesto Project estimates in terms of their face validity.