,
7 tweets,
2 min read
Read on Twitter
1. There are two issues with the approach: as others already mentioned, the way the items were combined causes problems with contextual/empirical validity. Second, the scale doesn't reflect the authors' own definition, resulting in issues of conceptual validity as well
#thread
#thread
2. As authors themselves note in Ch.7, the populism scale (anti-elitism and salience of corruption) doesn't reflect their definition (anti-elitism and invocations of vox populi). Why "populism" scale then? Especially as the operationalisation visibly impacts empirical validity?
3. CHES anti-corr item measures the salience of corruption as an issue, not broad rhetoric of "corrupt" ("distorted") vs "pure" ("genuine") politics. Thus, the way items were combined results in values cancelling each other out in some contexts. Not the fault of @ches_data
4. A gradualist rather than essentialist approach to measuring populism is nothing new. Beyond that and other points already raised by others, if a party is less "populist", it doesn't mean that it is automatically more "pluralist".
5. The overall critique is thus, to sum up, that for both empirical and conceptual reasons the chosen measure cannot be used to draw inferences on the extent of either "populist" (as defined by the authors themselves) or "pluralist" party appeals.
6. The points are not meant as nitpicking. I would also be the last who'd say that we shouldn't question what we already know. On the contrary, it's vital to continue working on our concepts and measures. My critique is hence not that the approach challenges extant insights.
7. In my view, it just doesn't really talk to them, particularly when linking concepts to empirical measures. Unfortunately, as this would have resulted in a much welcomed contribution for others to use and build upon. This is what we all want for our research field, in the #end
