Transitory and Resilient Salient Issues in Party Manifestos, Finland, 1880s to 2010s

Abstract

The performance of computational methods has been proven many times over. However, special efforts may be needed to ensure access to the research results achieved by means of these methods within specialized social science disciplines. This study joins previous efforts towards the mainstreaming of a specific computational method in the political science field of salience research. Rather than joining previous studies on the influence of the salience of issues to parties upon their electoral results or their propensity to form or join governments, this study represents the part of salience research that examines salience and its changes in their own right. Adapting ideas of digital historical humanities given the long study period, this study inserts salience theory within the frames of critical junctures theory to examine issues discontinuity, and path dependence theory to account for issue resilience. Using Latent Dirichlet Allocation topic modeling with 734 Finnish party manifestos from the 1880s to the 2010s as the research material, testing two hypotheses gave the following results. First, although many issues in the manifestos have been transitory, there have also been issues with resilience over critical junctures. Second, although there are resilient issues whose meanings have stayed the same by and large during longer periods, the meanings of some other resilient issues have pronouncedly changed, either suddenly at critical junctures or gradually during periods of path dependence. The implications for future studies are also discussed.

Artificial intelligence and voting advice applications

“…For instance, Lin et al (2015) proposed an automated text mining mechanism to classify legislative documents into predefined categories to allow the public to monitor legislators and track their legislative activities. Ahonen and Koljonen (2018) used topic modeling on a longitudinal dataset of party manifestos to differentiate transitory from resilient issues and to study whether the meaning of resilient issues has changed over time. Scarborough (2018) used supervised learning methods on Twitter to estimate attitudes toward feminism, concluding that Twitter could be a useful measure of public opinion about gender, although sentiment, as captured on Twitter, is not fully representative of the general population.…”