We assess different methods to identify party positions on immigration
What is it about?
In this article, we systematically assess different methods that can be used to identify party positions on immigration. We focus on immigration because this policy domain does not necessarily overlap with left–right, and because its salience and issue complexity vary across time and space. To assess the validity and reliability of different methods, we use 283 party manifestos (main parties in 8 Western European countries over 20 years) and different methods to obtain party positions: manual sentence-by-sentence coding using a conventional codebook, manual coding using checklists, automated coding using Wordscores, Wordfish and keywords. In addition, we use expert surveys and scores from the Comparative Manifesto Project (CMP). We find that the positions from experts, manual sentence-by-sentence coding, and manual checklist coding are consistent. By contrast, we find poor or inconsistent results with the CMP, Wordscores, Wordfish and the dictionary approach. An often-neglected method offers resource efficiency with no loss in validity or reliability: manual coding using checklists.
Why is it important?
Immigration increasingly transpires as an important policy domain to structure contemporary politics. With so many different methods, it can be hard to choose one. With no stakes in the game, we set out to compare different approaches. Automated methods like Wordscores or Wordfish cannot (yet) replace manual coding for all purposes.
Perspectives
http://dx.doi.org/10.1177/1354068817713122
The following have contributed to this page: Didier Ruedin
In partnership with:
