Weighting indicators is meaningless under necessary-and-sufficient concept structures

Evidence

Once there are multiple dimensions or indicators then the question of weighting arises—or should arise. Within an additive or statistical approach to measures this is not problematic per se. A statistical approach almost by definition gives different weights to different indicators. In contrast, it is not obvious what weighting would mean in a necessary and sufficient condition context or when one is using fuzzy logic to model the concept. For example, what does it mean to say that dimension 1 is “more necessary” than dimension 2? — Gary Goertz, Social Science Concepts, 306

Categorization

• SHARE: Nerdy tidbit for the one methods person in the reader’s life: ‘more necessary than’ is a category error, and additive indices smuggle in an assumption most concept users never notice.