Ideological Congruence and Social Media Text as Data

Abstract

Earlier studies on ideological congruence mostly rely on public opinion surveys to measure voter ideology, while politicians' ideology is measured by instruments such as roll call votes, expert surveys, and legislative texts. One crucial problem with such approaches is that the tools used to measure the elites' ideology are not identical to those used to measure the voters' ideology. The rapid growth of social media use offers a unique opportunity to directly examine the ideological overlap of elites and the electorate on a common platform using a common technique. This study examines over four million Twitter posts by legislative candidates from four major Turkish parties and their supporters between 2012 and 2016. After applying machine-learning algorithms to clean non-political content from the data, we employ Wordfish text scaling technique to extract the policy positions and compare the party positions to those of other parties and to those of their supporters.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.