What is it about?
Researchers increasingly use automatic approaches to coding party manifestos and other political texts. When these political texts are written in different languages, we might worry whether the results are comparable. Here I use political texts available in two languages to examine the impact of the source language on the results obtained. I use both the popular Wordscores method and a dictionary of keywords. When I use so-called stop words, the results are almost identical in both languages.
Photo by Jelleke Vanooteghem on Unsplash
Why is it important?
Automatic coding is a blessing to researchers, greatly reducing the work necessary to code political texts like party manifestos. With this, new research designs become possible, many of which are comparative in nature. The validity of these automatic approaches is central to their success. When working with different languages across countries, it is not evident whether differences in numbers are really differences in positions, or whether we merely pick up linguistic differences. This analysis reassures us that we pick up substantive differences.
Read the Original
This page is a summary of: The Role of Language in the Automatic Coding of Political Texts, Swiss Political Science Review, November 2013, Wiley, DOI: 10.1111/spsr.12050.
You can read the full text:
The following have contributed to this page